hackathon_id
int64 1.57k
23.4k
| project_link
stringlengths 30
96
| full_desc
stringlengths 1
547k
⌀ | title
stringlengths 1
60
⌀ | brief_desc
stringlengths 1
200
⌀ | team_members
stringlengths 2
870
| prize
stringlengths 2
792
| tags
stringlengths 2
4.47k
| __index_level_0__
int64 0
695
|
|---|---|---|---|---|---|---|---|---|
9,899 | https://devpost.com/software/htne-kitchen-sim | Scripting work by Daniel
Main Menu
Unreal Engine main level layout
Home Bakery
Start of Pizza Recipe
Pizza with sauce
End of Pizza Recipe
Learn to Chef
This is an entry for the Hack the Northeast hackathon (June 5-7, 2020) for the following categories:
Best Gaming Hack
Most Viable Startup
Created by Erik Marsh, Daniel Enriquez, and Colin Rogers.
Summary (Gaming)
This project was created using Unreal Engine 4.25.1, an advanced 3D rendering and game development environment.
Some assets were provided by the Unreal Engine Marketplace, and all others were manually created in the state-of-the-art 3D modeling software Blender and then imported to Unreal Engine.
Due to time constraints, the features listed below were not added, but would be present in a final product.
Simple drag and drop control scheme for interacting with ingredients and combining them to create a food item.
More recipes.
Recipe search based on ingredients.
Custom ingredient substitutions to accommodate for food allergens and intolerances.
Summary (Startup)
This is an application designed to teach people how to cook in a risk-free environment.
By risk-free, we mean that there is no risk of ruining ingredients by making mistakes.
Additionally, the user does not need to pay for ingredients, which will be useful for those in a low-income situation.
The target audience for this application is people in the age range of 18 to 25 with a desire to learn cooking skills that they have otherwise not learned, free of charge.
This application is designed to be approachable and easy to use for beginners.
Business Plan Outline
In the business summary we decided to go into detail about how a viable startup could be formed off the development and maintenance of this simulation.
Therefore, the business plan goes into detail about the marketing strategy, extra product details, and potential avenue for success if this were to be a startup that was made today.
A break-even analysis was also made to analyze the revenue produced by this simulation to ensure the viability of the startup with accurate figures for users.
The business plan mentions extra features that this team could not implement due to the limited time constraint. We ask the judges to look at the business plan as it details how a business can be made from this idea and shows the potential that a product such as this one has for growth.
Challenges faced
In such a limited timespace it was hard to implement the totality of the project, however, since we were able to implement the main components of the idea we feel as if we did an adequate job as we planned the viability of the startup on an actual idea and foundation that could be built. Another thing that was difficult was the business plan, we knew we had a viable startup and really wanted to go into detail at the same time because we felt the need that it was a very viable startup. Therefore, we took extra time to create a well detailed business plan and go above and beyond what was expected of us. This is a lot since we do not have any prior business knowledge yet still constructed a viable marketing strategy and business analysis.
The other key challenge faced was working with our primary tools, Unreal Engine and Blender are notorious for being complex yet packed with a lot of power. We believe that in the end, we did a nice amount of work with an adequate demo to show the capabilities that a kitchen simulator could show, as it has the potential to mirror your own kitchen.
More information can be found in the
Documents
directory.
How to run
Install Unreal Engine
Open
kitchen_sim.uproject
with Unreal Engine with the Kitchen Assetts pack installed
Built With
c
c#
c++
Try it out
github.com | Learn to Chef | A kitchen simulator to streamline learning how to cook and try new recipes without wasting ingredients | ['Daniel Enriquez', 'Erik Marsh', 'Colin Rogers'] | [] | ['c', 'c#', 'c++'] | 48 |
9,899 | https://devpost.com/software/envisionai-oriwv5 | Inspiration:
We were inspired by the various events happening in 2020, since we thought everything was going on so fast and seemed confusing to grasp, we though that it would be a good idea to analyze the trends of popular events evolving in countries
What it does
We utilized Angular.js, Express.js and Node.js to develop the frontend, and used Tensorflow API, python and Google trends api for backend and getting data
Challenges:
we had a hard time RNN machine learning and had to use Dense machine learning instead
Accomplishments:
We were able to complete a full stack app using machine learning that we have never used before
Built With
angular.js
css
express.js
html
javascript
node.js
python
typescript
Try it out
github.com | EnvisionAI | Trend analysis and predictor | ['Andy Yu', 'Ed Win', 'Simon Wang'] | [] | ['angular.js', 'css', 'express.js', 'html', 'javascript', 'node.js', 'python', 'typescript'] | 49 |
9,899 | https://devpost.com/software/college-major-finder | Thumbnail
Inspiration
There were 30 minutes left in the competition.
What it does
Assists you in finding a college major which fits your style and knowledge.
How we built it
Typing
Challenges we ran into
Slow typing
Accomplishments that we're proud of
Submitting
What we learned
We shouldn't have started so late
What's next for College Major Finder
Forgetting it
Built With
python
Try it out
github.com | College Major Finder | Take a quick trivia quiz and get some major recommendations! | ['Dagomara Miller'] | [] | ['python'] | 50 |
9,899 | https://devpost.com/software/blocktrace | Story
There is not a mainstream privacy preserving based contact tracing app.
We think that such a project would be vital for colleges to reopen in the fall.
Google and apple have shown interest, but their solution does not appear to be privacy preserving.
Technology
We used a block chain to be able to store the data decentralized, while also preventing the spoofing of any interactions.
This was written in c++.
We build a front end demo in Android studio which would be the user facing app, which would would track interactions without interupting privacy, allow the user to check if they interacted with anyone who has been diagnosed with Covid19 without knowing who they are specifically, and allow the user to mark that have been diagnosed to warn others.
How does it work?
The privacy preserving idea comes from that you send out randomized strings to phones that are within 6 feet of you and you keep track of all strings that you have ever sent. Then if you have Covid19 you will upload all of your sent out strings from the past 2 weeks into an "infected" string database.
Users can also check their list of people they have interacted with against the list of infected people in order to determine if they have interacted with anyone (this would be automatically done by the app)
This would allow determining interactions without compromising the privacy and even the location of a specific user.
Block chain allows for a prevention of spoofing as well as keeping the data in many different places, while making sure that it is perfectly up to date.
What have we done so far?
We have created several different aspects
We created the frontend using android studio, as well as the blockchain is built in c++.
Next Steps
With some more time we plan to integrate these different pieces together,
Move the blockchain into being fully distributed,
and pitch to universities as an important and smaller scale starting point for an app like this.
Built With
c++
cryptopp
http
sha256
Try it out
github.com | BlockTrace | Using privacy conserving contact tracing with a decentralized blockchain to maintain upmost privacy in contact tracing | ['Jordan Matthew Phillips', 'Suchetan Dontha', 'Abhishek Agarwal'] | [] | ['c++', 'cryptopp', 'http', 'sha256'] | 51 |
9,899 | https://devpost.com/software/providing-legit-job-postings-0jcvef | Team
We are a team of 3 consisting of Aerica Singla, Arun Venugopal and Arushi Madan.
Inspiration
COVID-19 pandemic is affecting economies in every continent. Unemployment rates are spiking every single day with the United States reporting around 26 million people applying for unemployment benefits, which is the highest recorded in its long history, millions have been furloughed in the United Kingdom, and thousands have been laid off around the world. These desperate times provides a perfect opportunity for online scammers to take advantage of the desperation and vulnerability of thousands and millions of people looking out for jobs. We see a steep rise in these fake job postings during COVID-19. In the grand scheme of things, what may start off as a harmless fake job advert, has the potential of ending in human trafficking. We are trying to tackle this issue at the grassroot level.
What it does
We have designed a machine learning model that helps distinguish fake job adverts from genuine ones. We have trained six models and have drawn a comparison among them. To portray how our ML model can be integrated into any job portal, we have designed a mobile application that shows the integration and can be viewed from the eyes of a job seeker. Our mobile application has four features in particular:
1) Portfolio page: This page is the first page of the app post-login, which allows a job seeker to enter their employment history, much like any other job portal/app.
2) Forum: A discussion forum allowing job seekers from all around the world to share and gain advice
3) Job Finding: The main page of the app which allows job seekers to view postings that have been run through our Machine learning algorithm and have been marked as real adverts.
4) Chat feature: This feature allows job seekers to communicate with employers directly and discuss job postings and applications.
How We built it
We explored the data and provided insights into which industries are more affected and what are the critical red flags which can give away these fake postings. Then we applied machine learning models to predict how we can detect these counterfeit postings.
In further detail:
Data collection: We used an open-source dataset that contained 17,880 job post details with 900 fraudulent ones.
Data visualisation: We visualised the data to understand if there were any key differences between real and fake job postings, such as if the number of words in fraud job postings was any lesser than real ones.
Data split: We then split the data into training and test sets.
Model Training: We trained various models such as Logistic regression, KNN, Random Forest etc. to see which model worked best for our data.
Model Evaluation: Using various classification parameters, we evaluated how well our models performed. For example, our Random Forest model had a roc_auc score of 0.76. We also evaluated how each model did in comparison to the others.
Further on, we designed an app titled JOB, using Adobe XD. We also integrated a Twilio framework so as to add a system to notify job posters whose job postings may have been flagged. This is because, since our ML algorithm is not a 100% accurate, we would have cases of 'False Positives' wherein a true job posting would be marked as fake. In these times we would want to allow the job posters to challenge the decision by providing evidence.
Immediate impact
Especially during but also after COVID-19, our application would aim to relieve vulnerable job seekers from the fear of fake job adverts. By doing so, we would be re-focusing the time spent by job seekers onto job postings that are real, and hence, increase their chances of getting a job. An immediate consequence of this would be decreasing traffic onto fake job adverts which would hopefully, discourage scammers from posting fake job adverts too. Police departments don’t have the resources to investigate these incidents, and it has to be a multi-million-dollar swindle before federal authorities get involved, so the scammers just keep getting away with it. Hence our solution saves millions of dollars and hours of investigation, whilst protecting the workers from getting scammed into fake jobs and misused information.
Revenue generated
Our Revenue model is based on:
1) Premium subscription availability to job seekers to apply for jobs
2) Revenue from the advertisements
3) Commission from the employers to post the jobs
Funding Split
1) Testing and Development: $ 10,000
2) Team Hire Costs: $ 6000
3) Patent Application Costs: $ 125 ++
4) Further Licensing conversations: $ 3000
TOTAL: $ 19,125
Future Goals
We would hope to partner up with LinkedIn or other job portals in a license agreement, to be able to integrate our machine learning model as a feature on their portal.
We also wish to completely automate the notification system built using 'Twilio'. Currently, it requires a command to run and is not automatically generated.
What we learnt
Through the medium of this hackathon, and with immense help from notebooks on Kaggle, Data Science blogs etc., we were able to create a ML program. It was new for us to code in JavaScript, but we developed our skills by learning through a need-based approach and were able to create a working Twilio program.
Built With
javascript
machine-learning
node.js
python
twilio
Try it out
github.com | Providing Legit Job Postings | Filtering out fake job adverts from existing job portals | ['Arushi Madan', 'Arun Venugopal'] | [] | ['javascript', 'machine-learning', 'node.js', 'python', 'twilio'] | 52 |
9,899 | https://devpost.com/software/steps4news | Steps4News
We help our users stay healthy and informed.
Quickly keep up to date with recent headlines while you exercise.
How it Works
Exercise sessions are securely backed up to the cloud.
Feature complete user account infrastructure.
Our Tech Stack
The Future
Get Steps4News Today
Inspiration
We feel there is a demand for a platform that combines exercise and keeping up with the state of the world, especially in the age of COVID-19.
What it does
Using Steps4News is intuitive. You simply enter a keyword and press start. Immediately, the app will start tracking your workout by counting the number of steps you've taken. In the background, you’ll hear a Text-To-Speech reader dictating the latest headlines and a short description for related news articles. After the workout, exercise sessions are then securely backed up to the cloud and stored in your account.
How we built it
Our tech stack consists of Firebase for authentication and database service. The application is written in Java and currently only supports Android. Information on the latest news is harnessed using the NewsAPI.
Challenges we ran into
Android's ecosystem is particularly hard to manage and test quickly. Some versions of Android do not implement newer APIs and we have to choose a version that made a compromise between market compatibility and features. Collaboration required the use of third-party plugs which frequently went out of sync.
Accomplishments that we're proud of
Successfully implementing Firebase Firestore and Authentication API in Android.
Sucessfully creating an Android ListView and integrating it with Firestore
What we learned
Integrating Android apps with Firebase's complement of services
Creating a ListView in Android to display a list of data (workout sessions in our case)
Retrieving data from firestore, firebase authentication and using it in the app
Creating sensor listeners to listen for android device's step counter sensors
Using Android's built in Text-To-Speech API to make it speak out news
Using Gson to parse API endpoint response data and displaying it to our app
No network activity on main thread! (caused us quite a bit of trouble) Threading is required to access network
Android development isn't as easy as we imagined
Zoom is actually quite a nice piece of recording software
Firebase has a ton of cool tools that we've never heard of before
What's next for Steps4News
Eventually, we want to be able to support listening to other forms of media such as podcasts in groups, leaderboards to allow friends to share and compete for total steps with one other, as well as integration with popular equipment providers and services such as Peloton or Google Fit. Monetization is viable using an advertising and subscription model.
We hope you find Step4News useful and is something you love to use and keeps you healthy, and informed.
Built With
android
firebase
gson
java
newsapi
Try it out
github.com
steps4news.web.app | Steps4News | We help our users stay healthy and informed. Get the most recent headlines while tracking your exercise. | ['Gary Li', 'Alex Yuan'] | [] | ['android', 'firebase', 'gson', 'java', 'newsapi'] | 53 |
9,899 | https://devpost.com/software/open-entex-cxuijm | Computer vision tracking
Foreground mask
Real-world CV testing
Data visualization
Firestore database
Inspiration
As a business owner, do you ever wonder how many people actually pass by the promotional sign you left outside your store? Or consider what might be the most profitable store hours for you to stay open? With the menacing threat of COVID-19, what if you could provide your customers with the comfort of knowing that your store won’t be crowded at the time they arrive? Imagine being able to count the people entering and leaving your store without needing to hire a human to do so!
As retail struggles to reopen in the wake of the COVID-19 shutdown, we designed open.ENTEX to help answer the aforementioned issues that businesses might be struggling with. Our goal is to provide small business owners with a tool that allows them to gather additional data for analysis and help optimize their businesses during these trying times.
What it does
open.ENTEX is a deployable computer vision application that analyzes camera footage to track people and their movement within the video frame. It is able to identify how many people walk by in each direction and count people entering and exiting a building. This data is then pushed to the cloud and can be viewed through the open.ENTEX-Web, our Svelte based Web-App.
Both business owners and customers benefit from open.ENTEX! Business owners could use our software to identify peak hours, potential sales, and marketing opportunities. Furthermore, during COVID-19 open.ENTEX can ensure that retail is adhering to guidelines and provide help with monitoring available capacity. Users can utilize the open.ENTEX web-tool to view the current number of people in a store and browse historical traffic data, helping them avoid crowds and rush hours.
How we built it & how it works
The open.ENTEX pipeline consists of three principal components: our custom Computer Vision algorithm, a Cloud Firestore, and a Svelte based Web App.
The Computer Vision Algorithm:
Using Python and OpenCV, our program processes a video feed and recreates a static background image using a technique called background subtraction. This allows our program to separate moving pixels from static ones and create a background image. The result is a generated foreground mask that filters the active pixels, leaving a predominantly black image with select white pixels to represent movement.
Next, we filter the foreground mask again and highlight subsections that pass a high enough threshold. However, the video still has excess noise due to the motion of shadows or changes in lighting. To resolve this, we implemented the number of islands strategy to detect large clumps of nearby neighboring highlighted pixels. This allows the algorithm to remove any loose pixels that were originally detected and isolate moving people. This data is then used to calculate the current location of each person and track their velocity. Using this information, we can determine whether a person was entering or exiting the frame. Finally, the data points are pushed to the cloud.
The Database:
Every time a person is detected walking in or out of the store, an update call is made to the database in Firebase. The database is sorted into collections and documents; each unique store is a document and has a collection of data points indicating the number of people in the store with a timestamp.
The Web App:
As we aimed to learn new technology, we decided to build our Web App using Svelte, an open-source JavaScript framework. When a user arrives at open.ENTEX-Web they are prompted to pick from the selection of store deployments. The data is visualized in three ways:
A store headcount
An “available capacity percentage bar”
A graph showing past and current data in relation to time
Data is loaded in parallel to all these components to increase speed for the user.
Challenges we ran into and what we learned
When developing the CV algorithm, our biggest challenge was identifying clumps of movement and tagging them as unique objects. We struggled to track these people over a series of frames, especially when an object disappeared only to reappear a few frames later. We were able to solve this by looking to see if any objects existed in past frames in a similar spot before disappearing and measuring the Pythagorean distance to decipher if it could be a realistic guess.
The main front-end difficulty was populating asynchronous data into the DOM of the Web App. We used Svelte, a newer alternative to Shadow-DOM based frameworks like React and Vue. However, none of our team had prior experience with Svelte-based data visualization. Being a newer framework, there are not nearly as many community-made and open-source imports. In the end, we decided to prioritize making a visually appealing and easy to use front-end instead of focusing on live updates. This would be an important feature to implement in open.ENTEX-V2.
Accomplishments that we’re proud of
First off, two of us are extremely proud of completing our first hackathon. Taking any compelling idea and compiling it into working software is always remarkably gratifying, and doing so in just under thirty-six hours is quite the achievement for any team. However, we are pleased that our project has many practical applications for both consumers and retailers. We hope that it can be used by the local community, and if it can have any impact during these troubling times, nothing would make us prouder.
What's next for open.ENTEX
There is only so much we could accomplish in thirty-six hours. In the future, we aim to improve both the computer vision and the user interface. While the CV is accurate most of the time, there are still a few edge cases in which it is slightly inaccurate, occasionally missing a person when people are too close to each other or are passing behind another object. The UI shows the data well and is easy to use. However, we lacked the time to add anything other than basic functionally.
From a business perspective, our next step would be to gauge interest with nearby retailers. This means coordinating test runs at the local supermarket and identifying the next steps before deployment.
Built With
firebase
javascript
opencv
python
svelte
Try it out
github.com
hacknortheast.web.app | open.ENTEX | The open-source computer vision web-app that offers small businesses and users easy data analytics about customer traffic. | ['Avi Porath', 'Ariel Simnegar', 'Alex Bulanov', 'Nadav K'] | [] | ['firebase', 'javascript', 'opencv', 'python', 'svelte'] | 54 |
9,899 | https://devpost.com/software/netra-59dgvf | Netra
Mono Mode (Nearby communication)
Fleet Mode (Anywhere communication between ffleetmates)
Setting Fleet Mode
Showing Restricted Area
Aerial Netra
Introduction:
Aerial Netra is a web platform that help drones transfer and receive data for higher security and better descision making. Basically aerial flight specially UASs or UAVs are much prone to accidents, very much due to unhealthy cloud conditions imprecise, detecting sensors and a lots of other reasons. And a lot of these happenings can simply be solved by transmitting surrounding data to them. And that's our solution, to suppy aerial vehicles with surrounding info from similar aerial vehicles around them. This can also help drone pilots to be aware od their drone's situation and surrounding to make descisions about the flight.
Basic Working:
Main feature basically is a relay mechanism that relays important variables like location-cordinates, temperature, pressure, altitude, speed, etc from a drone to nearest drones(vicinity can be choosen, for now 3 kilometers). This creates an information sponge for the drone which can help drone absorb surrounding information and act accordingly. For now we have two basic features or modes of operations for the platform:
Mono Mode
In this mode drone is alone and receives information from surrounding drones and transmits its information to them.
Fleet Mode
In this mode drone is accompanied by fleetmates so that they can share their info no matter where they are, to remain in contact and to aware of their fleetmates.
Activation of fleet mode:
Upon the press of switch button on the top-right corner of the page, adialog box is received. It contains Fleet-ID which can be used to transmit data between fleetmates, no mater where they are.
Technicals:
We used websockets to seamlessly transmit data between drones(which for now is our browsers and a poorly-coded python bot we created). We used Django as web-framework and Django Channels to make websockets work. For the front side, we used Vue-Js.
Story:
Prologue
We were thinking to do something with AR/VR with echoAR but then a short discussion on video chat drifted the idea to drones and then to our solution. We spent almost a couple of hours on topic research and planning and another hour on technical research jotting down packages.
Characters
Suraj Bhattarai started plucking out every piece we talked about and started defending his ideas. Saurav Niraula stayed quite in the ideation, only to start working like an ox afterwards. Sudesh Mate pointed out details and brought up technical spokes to support the project and Suraj Jha started connecting pieces to make them work.
Challenges & solutions
We didn't had hardware to do stuffs in the real way. So, we simulated browser as the drone and worked on it. Actually, noone had worked on websockets before and it was real pain understanding channels and groups and making them work through a series of "console.log"s.
In the nutshell
We started rapidly at first but then slowed in the middle due to issues with websockets and mapbox integration(mapbox was really and is), then again things went well and we satisfactorily pulled out our project.
Business Plan:
UAV's are damn inevitable. But still they are just at starting phase so we'll have to focus on drone pilots help and them use their drone out of the line-of-site. And because we'll be the very-early-mover, we'll have the advantage.
We can charge them on the volume of data received. That'll be our basic revenue model
In Team:
Saurav Niraula
Worked in frontend. Integrated MapBox and worked with Vue Js.
Sudesh Mate
Helped in research and worked on drone navigation.
Suraj Bhattarai
Worked on idea and documented the entire thing.
Suraj Jha
Created backend and helped to integrate it with Vue.
Implementation
We can't do this thing all alone. We will need government bodies like FAA to prepare necessary protocols and issue national permission upon which we can stretch our services
References:
Django
Django Channels
Vue JS
Channels Redis
Python Websocket Client
Mapbox
Built With
c
c++
css
gap
html
javascript
powershell
python
roff
shell
vue
Try it out
github.com | netra | UAV Reporting System | ['Suraj Bhattarai', 'SauravNiraula Niraula', 'Sudesh Mate', 'Suraj Jha'] | [] | ['c', 'c++', 'css', 'gap', 'html', 'javascript', 'powershell', 'python', 'roff', 'shell', 'vue'] | 55 |
9,899 | https://devpost.com/software/stick-man-in-photo-land | This is photo of our game in action!
This is an image before it is fed into Detectron.
This is the same image after it is being fed into Detectron.
Inspiration
The platformer genre is a staple among video game genres, and home to classic characters like Mario, Sonic, Kirby, Megaman, Crash Bandicoot, the list goes on. But there's only so many ways that new platformers can innovate upon the classic formula until people grow tired of the genre. So that begs the question: What's next for platformers? We believe the answer to that question is converting real world images to functional levels using Facebook's image segmentation API, "Detectron".
Instructions
To run our game, you must first launch our central manager and have a file named "HTNE_image.jpg" in the same directory as the game files. The manager will then call on a Google Colaboratory/Colab notebook and run Facebook's image segmentation AI. The program will run on its own, likely over the course of 30 seconds, but you will be prompted to put in an image. After you run the image through Facebook's Detectron AI, the program will output the original image and a version of the image that the game uses to make platforms. You must then download these images and put them in the same directory as the game files. Once both images are located in the game files the game is automatically launched in a Pygame window. This may take 10 seconds You then decide the locations of the Star to complete the level, the obstacles to avoid and the player spawn location! Use "WASD" to move, "Space" to jump, "R" to restart and "T" to toggle between regular image view and what the program sees.
How we built it
The entirety of this project was coded in Python and split into three programs. The first program is a central manager file that uses the os, keyboard and webbrowser python libraries to call on the other programs our directory. The second program was created in Google Colab and it calls on Facebook Detectron AI and runs the user-inputted image to do so. The last program is our game file that uses Pygame and calls upon the various images that we made and the Colab code generated to run. We designed the player sprites as well!
Challenges we ran into
The biggest difficulties we had when coding this project were opening and running Google Colaboratory from the Manager.py program, dealing with irregular shapes generated by the AI and producing viable stages using the Facebook Detectron API. We considered a variety of different additional downloads and libraries to run our program and make sure it was as user-friendly as possible but we ultimately concluded that the tools we used were the best for the job and we are very satisfied with the results!
Accomplishments that we're proud of
This project was filled with a lot of firsts! One of our members had never even used Python before Friday! From new libraries and new APIs to executing code in the terminal from IDLE and Google Colaboratory, there was a lot of new skills that we picked up for our project. So the fact that we managed to create a fully functioning game that works how we want it to means the world to us.
What's next for Stick-Man in Photo-Land
Saving, sharing and a streamlined photo import process.
Built With
facebook-detectron
google-colaboratory
pygame
python
python-package-index
Try it out
github.com | Stick-Man in Photo Land | The Platformer in Your Room | ['Josh Gole', 'Alex San', 'Devin Gulati'] | [] | ['facebook-detectron', 'google-colaboratory', 'pygame', 'python', 'python-package-index'] | 56 |
9,899 | https://devpost.com/software/recipe-1l7rab | Recipe Screen
Home Screen
Login Screen
Introduction
For a moment close your eyes and imagine. It’s sunday evening and you’re craving some mango fried rice. You search up a recipe and prepare your shopping list. As you shop at the supermarket, you realize that they don’t sell fresh mangoes. You are forced to take extra time and go to another market, not even sure if it has your wanted ingredients. How inconvenient is that?! Introducing Recipe, an app-based startup, that will prevent you from wasting easily avoided time, effort, and energy!
We use a combination of node.js, flutter, and puppeteer to create an app that can search for recipes based on the selected market location.
App
Users first create an account and logs into the app. With flutter’s efficent and easy integration with Firebase, we are able to easily implement Firebase features. We use firebase authentication to ensure that our user’s accounts and passwords are encrypted when they enter our server. This will also allow us to implement email/phone verification for accounts in the near future.
After they login, they will be greeted with a welcome screen, that will display stores nearby. This information is retrieved from the store collection in the Cloud Firestore database. They can then select a nearby supermarket that they have decided to go to. The app will then display recipes that can be made from the ingredients from that one selected supermarket. The app uses tracked current store inventory data, so the all needed ingredients are guaranteed to be available at the selected store.
Web-scraper
In order to create a database of recipes that relate to the inventory of each store, our nodejs program involves the use of web scraping and automation with puppeteer, and the use of a food recipe api, called spoonacular, and a cloud firestore database.
Puppeteer, a chromium automation plugin for NodeJS, first accesses the Safeway page, clicks all the load more buttons, and uses CSS selector to get the titles of the products.
We then insert the titles of these products into a findRecipeByProduct query with the Spoonacular API, Our implementation of this step is currently not perfect because of the fact that some branded product titles are not detected as ingredients. In the future we will use AI in order to find generic product names, given branded product titles.
Finally we insert the result of this query, along with other information about the store into the Cloud Firebase database, in the stores collection.
Future Plans
Our future plans mainly comprises of app development and bringing the app to market. We devised 4 stages towards the go to market plan to ensure startup success, each, lasting about a year. Stage A is to finish app development (expand store database to nationwide network, be able to find generic names of online product titles, use realtime inventory apis from stores) and create and to run pilot tests to make sure everything runs smoothly. Stage B is focused on promotion through social media and through sponsorships to increase our reach. Stage C is focused on increasing profit, where we will minimize our customer acquisition costs by streamlining our marketing efforts to the most effective platforms in past pilots, as well as limiting maintenance costs also from the data from pilot tests. Finally, stage D is focused on optimization, where we will optimize the previous three stages and prioritize efficient growth to lead a successful future.
Built With
axios
cloudfirestore
dart
firebase
firebase-authentication
flutter
getflutter
node.js
puppeteer
spoonacular
Try it out
github.com | Recipe | See all available recipes in a single supermarket, preventing you from wasting easily avoided time, effort, and energy! | ['Alex Gu', 'Tevin Wang', 'Leon Yee', 'Spencer He'] | [] | ['axios', 'cloudfirestore', 'dart', 'firebase', 'firebase-authentication', 'flutter', 'getflutter', 'node.js', 'puppeteer', 'spoonacular'] | 57 |
9,899 | https://devpost.com/software/util | Logo
Inspiration
This project is based off Economic Utility theory, an economic model that assumes every individual makes their decision based on how much satisfaction they can get. However, in real life situation, nobody has all the necessary information. With this project we aim to lower that information gap, introducing Util, an Economic Utility calculator.
What it does
Util is a purchase assistance app that allocates the budget to various commodities for the user to derive the maximum level of satisfaction. It can help prevent impulsive buying, reduce stress of having to chose within budget and prevent any sort of irrational purchase.
How we built it
We built the web app using Flask and Bootstrap. We initialize the database with few products. Looping through the data of the products, we try to find the satisfactions that user can get and apply diminishing marginal utility concept onto the calculation. Some products such as calculator has large diminishing marginal utility since there are no incentive to get a second one while some products such as banana has low diminishing marginal utility since person will love to have multiple bananas.
Challenges we ran into
We thought we could implement the program in an e-commerce environment, but that would take too much time. Instead we just made it a web app, where users can add items and change the data for each of them on their own. It is not possible for each user to have their own database for their products since we don't have a server that has large enough storage space. This is just a demo of a possible future application
Accomplishments that we're proud of
We managed to build and deploy a functional web app.
What we learned
Developing web applications in Python. Flask framework and the Jinja2 language, as well as some libraries related to Flask such as SQLAlchemy and WTForms. Deploying web applications with Heroku.
What's next for Util
Util can have many path ways, it can also be a mobile app and help mobile users to make decisions on what food they should buy or what items they should buy. It can even be a plug in on amazon or online grocery shopping, measuring combination of item that a person should choose. It can lead to a future that without needing us to make decision
Built With
bootstrap
flask
heroku
jquery
Try it out
github.com
secure-meadow-78220.herokuapp.com | Util | Imagine having a software to help you make decision. Imagine not having to bother with decision making. Our project Util helps you on that! It makes decision whenever you need to make decision! | ['Le Cherng Lee', 'Paweł Mazur', 'Kushal Sarkar'] | [] | ['bootstrap', 'flask', 'heroku', 'jquery'] | 58 |
9,899 | https://devpost.com/software/blurin8r | Real use-case from the Hong Kong protests
Test case from a previous hackathon
Running a blurred webcam capture on zoom
Inspiration
After watching a documentary of the Hong Kong protest and seeing the possible dangers of protesting online we realized that people's identity were at risk when they protested. The irony was that in many cases, the awareness the people were trying to bring by posting pictures and videos on social media were actually endangering the protesters on the front line more by exposing their faces. For example, in the case of Hong Kong protesters are extremely cautious of having their identity exposed, going so far as to cover their ears, back of the neck, tattoos, etc, anything they could be used against them in the face of the law. Though the media does sometimes censor the identity of protesters, when live footage is being streamed, it is often impossible to censor the identity of everyone in the background in real time.
What it does
Blurin8r is an app that processes a camera input and automatically blurs out faces. The output is then piped to a virtualized camera stream which can be used as a regular webcam for programs such as Zoom, Skype, or Facebook.
In addition, there is a web interface to edit the settings of the algorithm. The three settings we chose to be adjustible are the gaussian blur value, the blur sclaing factor, and the maximum size of the face. The gaussian blur value should be tuned to hide the identity of the face while still being smooth enough to see the background around the face. The blur scaling factor edits how big of an area around the face to blur, so the tops of the heads and necks can also be blurred out if necessary. Finally, the maximum size of the face can be used to isolate only the background faces for blurring. This is useful, for example, if the user would like their own face to be shown in the foreground while protecting the identity of those around him or her.
How I built it
The core algorithm has four main components: preprocessing, facial detection, blurring, and the virtual webcam.
Preprocessing
Before the image can be used, it must be preprocessed. To do this, it is first grabbed from an opencv video capture, then converted into a numpy array. The numpy array stores the pixel values in BGR format in a [X, Y, 3] array. Before running in the neural net, the image is then resized to a more manageable size and converted into grayscale. The resizing is necessary to reduce the number of necessary neurons on the input layer, therefore decreasing the overall size of the neural net and running faster.
Facial detection
The original method of facial detection that was attempted was using the stock haar cascade from opencv. While this worked on a few different use cases, it consistently missed faces. Any faces at an angle (profile view or tilted) or wearing sunglasses or a hat were not detected. Here is an example of a picture of my friends in Boston depicting a few missing detections:
While most of the faces are detected, there are a few missing. In our use case, we need consistent face blurring, as in a video, if one frame is missed, the privacy is lost.
To combat this, we used a neural network with the YOLO architecture. Here is the same image run through the trained neural network.
Originally, a YOLOv3 neural network was used, which resulted in a huge performance decrease (3FPS). After some research, we switched to a YOLOv3-tiny neural network, which had a near real time performance (20FPS).
The neural network weights were taken from
this Github repository
.
One thing to note about the facial detection is that this algorithm does not run facial recognition. This means that we do not identify who is in the background, we just identify that there is a face in that location. In addition, the original images are not stored, so the faces of those in the background stay hidden.
Blurring
Once the faces were found, a mask is created in the locations of the faces. The gaussian blur of the entire image is then merged with the original image using the
np.where
function, giving the effect of blurred out faces.
Virtual Webcam
Finally, to output the final image, a virtual webcam is created. We did this by using
v4l2loopback
, a dynamic kernel module which creates a virtualized camera capture source in
/dev/video20
. Then, from the python application, the final RGB values are piped into that source.
The reason why a virtual webcam had to be created was to allow compatibility with stock apps, such as Zoom, Skype, Twitch.tv, Youtube, or any other app that uses a webcam.
Challenges I ran into
We ran into a number of optimization issues with the YOLO neural network for face detection and blurring that prevent us from being able to stream real time video. However after performing some GPU optimizations and cutting the size of the neural network we were able to reduce lag significantly.
We also had trouble getting the server and neural network code to run simultaneously - after learning about threading, processes, and how they were implemented in python, we integrated a Queue and Processes into our main python code and got the server up and working.
Accomplishments that I'm proud of
We are proud to say that there was an great amount of learning done by all team members.
Personally I had little to no website development experience before this hackathon , so I came into this project with the goal of having a deeper understanding of HTML, CSS and Java Script and how they all worked together. I went through numerous iterations of the website design and going from a basic black text to the website we have now is really rewarding.
We took this hackathon as an experience to learn as much as possible in a short amount of time. We learned a lot about web development, neural network optimization, creating webcam streams and writing over them, and how to enable and work with multiprocessing and multi-threading within python programs.
What's next for Blurin8r
We completed a MVP demo of the core algorithm during the hackathon, but there are many steps going forward. The first and most obvious is to implement some form or GPU hardware acceleration. The current code runs at about 20FPS on my laptop using the
yolov3-tiny
neural net. Running with hardware acceleration such as CUDA, there is an expected performance increase of 300%.
In addition, running the demo on a laptop is not near the final application of this algorithm. A more useful application would be on cell phones, especially as it is the most common camera source. Finally, we can extend this to run on the original intended platform: news cameras.
Read more
See our slide deck here:
https://docs.google.com/presentation/d/1vpuOZA1IcWGVWcdjG47TVGtR2mzEhbXLzh_SPM77LDw/edit?usp=sharing
Built With
css
flask
html
javascript
opencv
python
yolo
Try it out
github.com | Blurin8r | Keep your face to yourself | ['Emily Wan', 'Wesley Soo-Hoo', 'Adi Ramachandran'] | [] | ['css', 'flask', 'html', 'javascript', 'opencv', 'python', 'yolo'] | 59 |
9,899 | https://devpost.com/software/body-mass-index-calculator-7sw54y | Inspiration
I'm an undergrad computer science student studying at the University of Alberta and I'm highly interested in exploring, learning and I'm motivated to build more projects, hence, I started with this!
What it does
This is a body mass index calculator which only requires your weight and height and is able to tell you your BMI and whether you're under weight, normal weighted, over weight or obese!
How I built it
This is honestly my first time ever building an app or first time participating in a hackathon for that case!
I learned how to use Kivy and the KV language in a matter of few hours and was able to attempt to create this app.
Challenges I ran into
When I reached the part where I had to display the BMI on my screen, I ran into trouble. I was able to display it on the python shell but was unable to display it on the app.
Accomplishments that I'm proud of
I found out about this hackathon a day ago and I'm glad I was able to learn so much on the way and grow as a person and as someone in the field of computer science!
What I learned
Even though I hit a brick wall, I'm still motivated to develop more apps by either using Kivy, Tkinter or other such apps. No matter how many times I fail, I know that nothing is going to beat the happiness I feel when I accomplish my goals. :)
What's next for Body Mass Index Calculator
I wish to further develop it by adding in categories such as body fat % and body measurements to create a more accurate app that calculates BMI.
If possible, I would love to receive feedback about my project and tips regarding how to proceed with building projects so I can learn and add valuable things to my resume!
Built With
kivy
pycharm
python
Try it out
github.com | Body Mass Index Calculator | Keep your health in check by knowing your BMI! | ['Jaspreet Kaur Sohal'] | [] | ['kivy', 'pycharm', 'python'] | 60 |
9,899 | https://devpost.com/software/headsup-gqkl3s | Python Tkinter GUI screen
HTML Version Home Screen
Live graph
Inspiration
In today's society, over 40% of children are diagnosed with bad posture. Every person is either on their computer, laptop or phone, looking down at their screens. The COVID-19 pandemic has exponentially increased the computer usage for both adults and students. This leads to great amounts of stress in the shoulders and can lead to health problems. This applies to both the school and corporate environments. To help solve this problem, I created this app that alerts a user when they are maintaining bad posture on their laptops.
What it does
It alerts a user when they are maintaining bad posture on their laptops by blurring their screen, dimming it, and playing an alarm.
How I built it
Using visual recognition, HeadsUp recognizes the different features of your face and measures the change in the nodes set at activation. I used the Tensorflow.js PoseNet model to recognize the posture on HTML and used OpenCV in Python.
Challenges I ran into
I was not very familiar with how to create a GUI with Tkinter in Python. This caused me some trouble; although the GUI is not as advanced, I had no idea how to use it. Alongside that, the training of my visual recognition model was failing periodically for unknown reasons which proved to be a setback.
Accomplishments that I'm proud of
Now I can use Python to create apps with a GUI. I am also proud of learning different Git commands which helped me with repository work.
What I learned
Persistence is the key to accomplishment!
What's next for HeadsUp
Implementation into everyday environments, such as in offices to provide possible healthcare benefits, analyze student engagement in classrooms, and later be able to diagnose certain diseases with data from this app. Not only that, but a mobile version that checks both screen time, and how close someone is to their phone.
Built With
css3
html5
javascript
keras
ml5.js
opencv
p5.js
python
tensorflow
tkinter
Try it out
github.com
shivamsyal.github.io
docs.google.com | HeadsUp | A web and computer app which alerts users when they are maintaining bad posture, using visual recognition and machine learning. | ['Shivam Syal'] | [] | ['css3', 'html5', 'javascript', 'keras', 'ml5.js', 'opencv', 'p5.js', 'python', 'tensorflow', 'tkinter'] | 61 |
9,899 | https://devpost.com/software/seeway | Intro page 1
Intro Page 2
Home Page
Login Page
Red Zones related to Covid-19
Commodities Monitoring with sensor value as well as as current geographical position.
Is Driver sleeping? What is the speed of the vehicle?
Upload Medical Document to see if you are postive or negative
Items Details
Truck Details
Manager Dashboard
Truck Driver Dashboard
Dashboard for Managaer
Inspiration
Did you know that the average
cost per truck accident is around $14,000?
That is a huge amount of loss!
Another fact that really struck us is that up to
53% of truck drivers reported physical & psychological issues
- which inadvertently leads to a higher chances of accidents.
During transportation of goods, we also realized that the loss incurred due to the
damage caused to the products is at least 150% of the price of the goods being shipped.
What a quite extravagant amount of money to be borne by a company!
And in our current state of a global pandemic, it is crucial to ensure the flow of essential goods to all. But did you know that in our hometown of Tamil Nadu in India,
out of 6000 cases recorded, 15,00 cases were linked to a vegetable market!
People are scared to now go get essential and the most basic necessities- fruits & vegetables!
These facts really made us think on how we can employ technology to create a wholesome solution that helps retailers/suppliers/traders to run their businesses better and safe.
That is why we have come up with our product -
Seeway
. We really want to help businesses - especially, the local suppliers/traders in India who have not yet witnessed a digital transformation in their business
We strive to empower our customers build the right future. We aim to be your partner in ensuring the safety of your people as well as your goods. By joining us, you have taken the first step towards understanding your employees as well as ensuring safety for your customer.
After all, the current situation made it crystal clean to all of us of what actually matters in life and food/supplies is definitely top of the list!
What it does
Our product - Seeway - is a holistic system that allows drivers and managers(retailers/suppliers/traders) to manage and control their day to day work of moving goods from one place to another.
It has the following modules:
Accident Prevention
- We install a webcam in the dashboard of the truck as well a proximity sensor. We use the datapoints(incl of speed of truck) to figure out how likely it is for a driver to meet with an accident. We used OpenCV and Keras API to build our ML model and Thingspeak for capturing data from our IoT sensors. An alarm will be sent to our app in order to wake the driver up and also notify the manager.
Commodity Monitoring
- We install a humidity/temperature sensor, shock sensor, vibration sensor, & smoke sensor in the back of the truck where the commodities reside during transport. We use the data points from these sensors and analyse them to ensure that the commodities are in the right environment settings. If not, an alarm will be sent to the driver through our app as well as a notification sent to the manager.
Driver Health Check-IN
- Drivers can now upload a medical document containing results of their illness , in our case due to the current situation, COVID-19 tests. The image will be uploaded into our database where Google will then extract text and see if the particular driver has been tested negative or positive. Accordingly, the drivers need to constantly upload medical results every month so as to ensure safety for him/her as well as others. The manager can have an overview of all the health statuses of his/her drivers.
Informed Maps
- Drivers and Managers have overview of a map in the app that shows the red zones. There is also information regarding the state tax he/she is in during transit. There is also overview of the distance covered and time duration to reach the destination.
Analytics Dashboard:
- Managers have an overall view of the trends seen in the environment of the items in transit as well as drivers in a comprehensive web app.
Currency Rates
- Managers can easily see the trends of the economy as well as the present currency rates.
Customer Base
- The app has also links to third-party market places like Ali Baba, Market India & so on - these links are helpful for local traders/suppliers using our app since most of them are not aware of the popular online market places out there.
How we built it
We primarily used Android Studio to develop our Android App.
We used Firebase as our real-time database to store all our data points.
ML-Kit was used for text extraction and Keras API was used to build our ensemble model.
OpenCV was used for Image processing of video of driver to see if he was sleepy or not.
Dash was used to develop our Health Dashboard that shows overview of data to Manager.
Material UI was used to design our interface.
Thingspeak cloud was used to receive our data from sensors and process streaming points.
Challenges we ran into
Hardware components were really hard to get due to the COVID-19 situation. But we have written the codes nevertheless with the sensors we had.
36 hours was a tight deadline for us - since we were doing it virtually and also both of us are 1000s of miles apart - attaining virtual collaboration in the first day was hard.
Much more smaller challenges here and there - but most importantly, we came through it all!
Accomplishments that we're proud of
Getting our BlinkRate detection model working was a great feeling! OpenCV is truly awesome!
Using Firebase ML-Kit was an intriguing experience since we had never used it before - using the text extraction was hard at first but then we got it to work.
Meeting a wonderful bunch of people! :)
Learning as much as we could from the mentors who were kind to guide us through everything.
What we learned
One of the key takeaways from this experience is us realizing that local business really need to be empowered by technology. We really understood the problems in essential businesses run in India today. Especially by those who are local such as the kiranas.
One of our team member's (Sakthisreee) grandfather is a chillies wholesaler and we had talked to him to get idea on how we could empower his business. One thing to know before moving ahead - traders like Sakthisree's grandfather don't possess a computer or use any other advanced medium in their stores. Also, the drivers they employ come in sick due to chillies-based illnesses and they get overworked a lot. Yet, so far, these businesses have run as usual, but with a certain amount of disadvantages. Taking such a situation, imagine what would happen if we gave a technological make-over for all such businesses! It would be great for them as well for the country's economy.
Apart from this, we learnt a lot technically - how to use ML Kit, get an Android App up & running and also, cool stuff like pitching!
What's next for Seeway
From what you read before , it is clear that we really want to give opportunities for the local traders and suppliers to utilize tech in their business. We aim to further improve our platform and app and add more comprehensive functionalities - adding in native languages as well apart from English.
Our next upgrade would be to make a digital twin of the supply chain operations so that the process is mapped from beginning to end and exclusive control is given to the managers in charge. Digitization also results in data - this data can be used for analytics and looped intelligence.
Built With
android-studio
arduino-sensors-=>-iot-android-mobile-app-thingspeakiot-firebase-analytics-firebase-mlkit-for-ocr-recognition-gsheet-cloud-database-machine-learning-(randomforest
dash
firebase
flask
keras
ml-kit
opencv
python
thingspeakiot
Try it out
github.com
drive.google.com | Seeway | We are your partner in ensuring the safety of your people as well as your goods. | ['Sakthi Sree', 'Anupam Rajanish'] | [] | ['android-studio', 'arduino-sensors-=>-iot-android-mobile-app-thingspeakiot-firebase-analytics-firebase-mlkit-for-ocr-recognition-gsheet-cloud-database-machine-learning-(randomforest', 'dash', 'firebase', 'flask', 'keras', 'ml-kit', 'opencv', 'python', 'thingspeakiot'] | 62 |
9,899 | https://devpost.com/software/ism | We were inspired by the current events going on in the world that led us to build this app. Instead of Instagram where all these types of posts would be cluttered and the information wouldn't be as effective.
What it does
We built the app using node.js with a react-native framework. We also used firebase authentication as well as firestore to create different screens and create an authentication system.
We learned about the different abilities of cloud firestore as well as the firebase database and how powerful it was.
Built With
firebase
node.js
reactnative
Try it out
github.com | ISM | Our product is a platform that allows people of all ethnicities, genders, and races make comments on current world situations and how they personally feel about them. | ['Tejas Polu', 'Arush Medam', 'Cailean Fernandes', 'Yash Chhatre'] | [] | ['firebase', 'node.js', 'reactnative'] | 63 |
9,899 | https://devpost.com/software/vibe-check-q93jaz | Landing Page For User
Example of Spotify Login
Personal Playlist Selection
Public Playlist Search / Selection
Vibe Check Page Before
Another Hover Example
Vibe Check Page After
Hover Over Text For Explanations
Inspiration
Whether you’re in the zone or just jamming out, a poor song choice can pull you out of your mood and out of your flow. According to researchers at University of California, Irvine, it takes 23 minutes and 15 seconds to refocus after an interruption - by ruining vibes and losing valuable time, it’s no wonder that poorly curated playlists were such a common frustration across our team. In order to solve this problem without spending hours on end cleaning through old playlists, we hoped to build an easy-to-use online tool that could help music lovers clean their playlists to their liking.
What it does
Vibe Check allows Spotify users to curate existing playlists both public and private to maintain a consistent “vibe” throughout the playlist. Users can access their own playlists as well as search public playlists to find playlists with the right overall vibe. Once they’ve found a suitable playlist, users can set the sensitivity of the vibe checker and use our vibe checking functionality to generate a new, curated playlist, removing tracks that stray too far from the overall vibe of the original playlist. Additionally we trained a neural network that is able to classify the “vibe” of the playlist through over 1000+ songs of training data.
How we built it
We began by exploring what information we could access from the Spotify API. Once we knew what data we could use, we brainstormed and storyboarded the user experience in Figma, which helped us decide the general logic, tools, and data structures to use while building the project. We then translated our Figma storyboards into an HTML and CSS frontend and connected the backend with node.js. After completing our core vibe checking functionality, we used brain.js to build and test neural nets to use for our vibe classification features.
Challenges we ran into
This project was certainly challenging as it was made with languages that our team had very little experience on. Firstly, our team was composed of backend developers which proved to be difficult as we needed to implement a usable frontend to users in order for the backend to be utilized properly. The creation of the neural network took quite a while as no one had used brain.js nor had experience with machine learning in the past. The amount of knowledge it took to correctly train a neural network certainty was a challenge and there were many trained neural networks that failed or couldn’t even predict between two entirely different genres. The use of the spotify API also had presented many challenges as the AJAX calls often gave us errors with rate limiting and also sync issues that our team had never encountered before. However, although we were using languages that were very new to us with libraries that we had never seen before we managed to build the project to a level of completion that we are extremely proud of.
Accomplishments we're proud of
As a team we were proud of many of the achievements that we were able to make during these 36 hours. We strengthened our interpersonal skills through working with teammates online even when time zones didn’t match up and also communicated effectively not wasting any time while coding and planning. Additionally, we gained many hard skills and improved in areas that we had not worked on much before in the past. We each went outside of our comfort zones and utilized new tools. With very little experience in the roles we chose we pushed ourselves to learn and apply our knowledge in real time and were able to create a web app that we would use even outside of this hackathon. We were able to connect all these pieces scattered around and tie all the concepts up into one package which we all found to be incredibly satisfying to see all our hard work come together in the end. We were able to come out of this hackathon more confident about our skills and are able to have something to show for all our hard work in this time, effort, and care intensive 36 hours.
What we learned
While building Vibe Check, we learned a lot about web development, API calls, machine learning, Javascript, HTML, and CSS. Each team member had very little experience with what they worked on making this an extremely learning intensive project. Our team members were able to get a solid grasp on creating the frontend for web apps through learning HTML, CSS, and Javascript to create the UI. In the backend we were able to learn more about data manipulation. This allowed us to transform seemingly meaningless float values into a valuable filtering tool for any song. We learnt about standard deviations, Z-scores, and how to implement these concepts in a real life application. Additionally, we were able to set up our first nodes with node.js and also host our first website using heroku. Lastly we were able to learn about machine learning by training our own neural network by learning about XOR gates, forwards propagation, backwards propagation, and also learning rates for the neural network. Overall, we gained a large amount of new knowledge and developed our skills to the point where we were able to make a cohesive project exceeding our expectations.
What's next for Vibe Check
In the near future, we plan to improve Vibe Check by:
Training our classification system with larger datasets more representative of the distribution of music on Spotify. We hope to make more accurate predictions this way, and we’re also exploring other classification algorithms.
Fine tuning the vibe filter so that the position on the slider will be more accurate and correspond to a more intuitive feel on which songs will be removed. Currently, we’re assuming that track features are distributed normally, but the distributions of particular songs are often hyper dispersed.
Improving the UI and UX to allow users to have a smoother experience while using Vibe Check. We want to allow users to keep specific songs slated to be removed for a more customized experience.
In the end we want to ensure our platform can prevent users from losing their vibes in their next workout, study break, or personal jam session.
Built With
brain.js
css
figma
heroku
html
javascript
node.js
spotify
Try it out
vibe-check-htne.herokuapp.com
github.com | Vibe Check | Never lose your vibes again. | ['Johnson Su', 'Andrew Yu', 'Amy Han'] | [] | ['brain.js', 'css', 'figma', 'heroku', 'html', 'javascript', 'node.js', 'spotify'] | 64 |
9,899 | https://devpost.com/software/neuros-mjyqx7 | Welcome and enter app page.
Page to add a person.
Matching memory game with score and example icons.
We were inspired by the memory loss that people affected by conditions of dementia face. It starts out with a slow cognitive decline which can mean forgetting the name of an object, but it can lead to mixing up and even forgetting loved ones. This is something no one should have to face and we wanted to find a way to solve this problem. Our app is called Neuros, and it is an app to help with forgetting people, objects, and general cognitive decline. The main feature of our app is adding and recognizing a person. Users can take a picture of someone, tag them with a name, and add memories that they have had with the person. This information then stores locally and can be used later to identify people. If a user forgets who a person is, they can take a picture of them and they will be taken to an area where it shows the picture of the person, their name, and the memories which can help trigger a pathway in the brain to help the user remember who the person is. Additionally, users can take pictures of objects and be alerted with what the object is and the definition of the object's name. This was all made possible by machine learning from Microsoft Azure. There are also 2 games implemented in the app. There is a matching memory game where users must match two icons to gain a point, and there is a trivia quiz game where a user is shown a picture of someone from their local storage and they must answer with the name of who they are. We learned a lot about the everyday challenges of having a condition of dementia. The problems that people face are very hard to overcome, and through our research about how to treat it we were shocked to see what many people have to go through. We only hope that our app can in some way provide a light to the condition and help treat the condition and ultimately, raise awareness.
Built With
azure
expo.io
javascript
react-native
wordsapi
Try it out
github.com
docs.google.com | Neuros | Never forget the people in your life. | ['AnirudhAdiraju Adiraju', 'Rahul Gupta', 'danielzhuang11 Zhuang'] | [] | ['azure', 'expo.io', 'javascript', 'react-native', 'wordsapi'] | 65 |
9,899 | https://devpost.com/software/gyft | Gyft
Empowering every community and small businesses to be financially agile.
GitHub Repository Breakdown
The Github repository is separated into two Flutter apps.
htn2020: This is the Flutter app with functionalities such as Log In, Sign Up/Register, and Querying data from the Firebase. Essentially, this is the backend component of the Flutter app. This is prototyped on Android (as half of our team used Android simulators).
gyft-1: This is the Flutter app with the UI interface coded up as seen from the Figma wireframe. Functionalities include navigation/routing, buttons, and input data. This is prototyped on iOS.
Links
Video Demo/Presentation
DevPost
Figma Wireframe
Slide Deck
Business Plan
Inspiration and Problem Statement
The COVID-19 pandemic has severely disrupted how small business across the US operate and generate revenue. Many small businesses are struggling to stay afloat in this economy. Many leaders and media outlets have advocated for purchasing gift cards from small businesses in an effort to alleviate these pressures. As Forbes put it, "buying a gift card puts money in the hands of a small business immedaitely."
However, Gift Card Systems are difficult to set up, and costly to maintain. In addition, many community members don't know that their local businesses offer gift cards, and purchasing gift cards requires effort and in many cases, in-person interaction.
This presents a unique challenge and opportunity space to be addressed.
Purpose and Value
Gyft is a platform that empowers communities by connecting local businesses with community members and potential customers. It allows businesses to start selling online Gift Card/Certificates for their business quickly, intuitively, and affordably, even without past experience.
Gyft enables small businesses to be financially agile during tumultuous times of crisis (such as during COVID-19). The ability to generate revenue even during shutdowns has been reserved for large, corporate businesses in the past. Gyft seeks to change that. Moreover, Gyft continues to add value to businesses beyond these times by generating revenue, attracting new customers, strengthening customer loyalty, and boosting branding.
Gyft also empowers community members to continue to patronize and support their local businesses, even when there is a temporary shutdown due to coronavirus. It also draws family and friends closer together through gifts for special occasions.
How to Use
See Slideshow for Image
Sign up/Registration
Account Confirmation
Select Category
Explore Businesses in Your Community
Customizer Gift Card
Purchase Gift Card
Use Gift Card
Marketing and Audience
Currently, there is no alternative solution on the market that makes it easy and affordable for small business owners to create their own gift card program targeting their local communities. Furthermore, it is difficult for community members to discover small businesses in their area that need their support. Gyft fills in this gap in the market.
Gyft’s target audience are small businesses and community members. Especially during times of crisis and shutdowns (COVID-19), there have been calls by leaders and the media to purchase gift cards to support local businesses [4]. However setting up a gift card program is a long and tedious process.
Our marketing strategy includes advertising through online websites targeted at small business owners and COVID-19 related content. In addition, there will be social media advertisements to users in areas where there are small businesses in need of support. Finally, small businesses will generate awareness of their gift cards on the Gyft platform for themselves, which adds to our marketing efforts.
Pricing
Revenue will be generated by charging a one-time 3% service fee on every gift card purchased. In other words, if a customer purchases a $50.00 gift card, the business will receive $48.50. There are no monthly fees or other charges.
Our fees are highly competitive compared to a physical gift card program. It can cost businesses $300-$400 for gift card software, and over $2 per gift card. Transaction fees and lengthy, restrictive contracts add even more costs [1].
Market and Financial Potential
According to the US Small Business Administration, there are 30.7 million small businesses in the US in 2019 [2]. Of these, over 2.6 million are in Retail Trade, and over 0.9 million are in the Accomodation and Food Services Industry. These two industries are examined because they are highly suitable for the Gyft platform. Small businesses in these industries generate more than 345 billion dollars of profit annually [2]. Gyft seeks to absorb some of this revenue, and generate additional revenue streams beyond existing in-person channels. In 2016, consumers spent 27.5 billion dollars on gift cards during the holiday season alone [3]. Gyft will tap into this market by providing consumers an alternative to gift cards that are dominated by large, corporate chains.
Technology
The Gyft platform consists of mobile applications (Android and iOS) built using Flutter and Firebase/Google Cloud Services. In developing the mobile application, we also used Figma for wireframing.
The main back end model was abstracted as an inheritance infrqastructure. So the children of this inheritence model all had a common thing i.e. they had a collection of Gift Card type object. This made it very easy to code in flutter and firebase as each collection was also a collection in the firebase database. The modularity if having a Gift Card Object was that the object could also have it's own section in the database and that made it really easy to parse in Dart Code.
In layman terms, the main focus of the project was making as much use of the database or cloud so as to make the infrastructure on front end as little as possible. Flutter was really helpful as navigating to different screens supprted our development intent. The good news was that we did all this 36 hours and got a great MVP to showcase.
Firebase stores consumer credentials, and participating businesses list is populated after contacting Firebase.
Challenges we ran into
Challenges we ran into fell into two large groups:
Unfamiliarity with the codebase: For all of the members, this is the first time developing with Flutter, as such, it took some time to get used to the structure of program and to implement the features we wanted to create.
Compatibility issues: Because half of our group used Android and the other half used iOS, we were unable to combine the UI interface (coded in iOS) and backend functionalities (coded in Android). As such, we ended with two different codebases with not enough time to merge them together.
Accomplishments that we're proud of and what we learned
We are very proud of being able to working together and delegate different tasks efficiently. Even though the codebase was new to us, we were able to create a coherent project and pitch it effectively. Through this endeavor, we learned a lot about mobile development, both in terms of creating a captivating UI and interacting with servers and databases.
What's next for Gyft
For the next steps, we hope to merge functionality with UI for our app and further hone the small mistakes that still exists in our current iteration. We also want to market our product to investors and consumers so that we are able to empower every community and small businesses to be financially agile.
Sources:
[1]
https://www.costowl.com/b2b/gift-card-custom-program-cost.html#:~:text=Depending%20on%20how%20many%20gift,%24300%20to%20%24400%20per%20location
.
[2]
https://cdn.advocacy.sba.gov/wp-content/uploads/2019/04/23142719/2019-Small-Business-Profiles-US.pdf
[3]
https://www.softwareadvice.com/resources/gift-cards-for-small-business/#holiday
[4]
https://www.npr.org/2020/03/20/818797729/how-buying-a-gift-card-can-help-a-small-business
Graphics and Image Credits:
https://stories.freepik.com/illustration/bankruptcy/pana
https://stories.freepik.com/illustration/queue/pana
https://stories.freepik.com/illustration/online-shopping/pana
https://stories.freepik.com/illustration/make-it-rain/pana
https://www.theverge.com/interface/2020/4/14/21219289/apple-google-contact-tracing-app-android-ios-pros-cons-quarantine-testing
https://www.underconsideration.com/brandnew/archives/android_2014_logo.png
https://www.google.com/url?sa=i&url=https%3A%2F%2Fwww.developersforhire.com%2Fdatabase%2F&psig=AOvVaw0wVW7u-nAr7kmQRUg-_omk&ust=1591629511995000&source=images&cd=vfe&ved=0CAIQjRxqFwoTCMj-g5OA8OkCFQAAAAAdAAAAABAg
Built With
android
android-studio
dart
figma
firebase
flutter
gcp
intellij-idea
ios
Try it out
github.com | Gyft | Empowering every community and small businesses to be financially agile, through easy and affordable gift cards. | ['Alex Xu', 'Achilles D.', 'Franklin Boampong', 'Steven Cheng'] | [] | ['android', 'android-studio', 'dart', 'figma', 'firebase', 'flutter', 'gcp', 'intellij-idea', 'ios'] | 66 |
9,899 | https://devpost.com/software/opinion-storm | Opinion Storm Homepage
Opinion Storm Logo
Opinion Piece Submission
Comment Submission
Inspiration
My platform serves as a combination of the New York Times and Twitter. I was inspired by Twitter's ability to allow random people all over the world to share their ideas with the world. However, I found that Twitter's character limit prevented people from sharing anything of substance. I was also inspired by how the New York Times published Opinion Articles from people across the country who had something to say. However, the selectivity of the New York Times prevented the vast majority of people from sharing their ideas, as most articles were simply rejected. These two platforms and the challenges with the platforms inspired me to create Opinion Storm.
What it does
Opinion Storm provides a home for opinion pieces written by normal people. The site allows anybody to post an opinion piece on the website and allows people to read the opinions of their peers. After reading an article, users can post comments and decide if they agree or disagree.
How I built it
I built Opinion Storm using Wordpress. I installed plugins and customized the plugins and the theme to create a web service that was useful and functional. I preloaded content under creative commons licenses from The Conversation.
Challenges I ran into
One challenge I faced was allowing users to submit content, as that was one of the core features of the project. I did some online research and was able to install a plugin and make adjustments to allow users to submit articles through the front end. I am glad I overcame this challenge, as this feature is fundamental to my project.
Accomplishments that I'm proud of
I am proud of creating what may become a startup. I am glad that I was actually able to follow through with my goal and create something functional and useful for the world.
What I learned
From this project, I learned quite a bit about HTML. I also learned a bit about domains and web hosting.
What's next for Opinion Storm
Next, Opinion Storm will feature advertisements, creating revenue, and ensuring that opinions will have a permanent home forever.
Built With
css
easywp
google-drawings
html
like-dislike-counter
newsium
user-submitted-posts
wordpress
Try it out
opinionstorm-6979cb.ingress-bonde.easywp.com
github.com | Opinion Storm | Opinion Storm is a platform that allows users to read and write opinion pieces, providing the freedom to persuade and be persuaded. | ['Alex Shieh'] | [] | ['css', 'easywp', 'google-drawings', 'html', 'like-dislike-counter', 'newsium', 'user-submitted-posts', 'wordpress'] | 67 |
9,899 | https://devpost.com/software/thecookingbot | Full view of IoT: 2 ultrasonic, 1 temperature/humidity sensor
IoT solution in action
Various statistics and warnings regarding cooking
Humidity data and pot overflow data warnings.
Timer stops to mark notes through cooking process
Add notes to recipes
Select recipes
Stored timed recipe data
The Cooking Bot is an inexpensive yet powerful device that aids as both a Cook Book and kitchen safety device.
With a multitude of sensors that monitor temperature, humidity, and distance, the Cooking Bot sends you a notification
if you are away from the cooking station for too long, detects an alarming drop in humidity or alarming rise in temperature,
or if it detects liquid overflow in your pots and pans! In addition, our unique timestamp|temperature software
allows you to create recipes with careful, reproducible accuracy. The notion of a "bad cook" would become delightfully impossible!
Not to mention, these extraordinarily precise recipes are in a compressible, easy-to-use format that makes sharing across many platforms as easy as pie!
Inspiration
Our parents (mothers) often scold us for not appreciating home-made food while we still have it. We wanted a way to enjoy the delicacy
of home-cooked food from our own kitchens, while also encouraging our more experienced counter-parts to be more mindful, and responsible
in the kitchen. According to the NFPA, cooking was the leading cause of reported home fires and home fire injuries. With the ideas of
protecting those we love and spreading happiness in mind, The Cooking Bot was born.
How We Built It
Hardware:
We wired together an ESP8266 with ultrasonic sensors, a humidity sensor, and a temperature sensor
in different directions to get many different datapoints at once. We used C++ to code the arduino that served as the controller for the
ESP8266 and sensors.
Software:
Both the front and backend were done with React-Native and Node.js, but all the sensor data was managed through firebase and
firebase APIs. We used Expo CLI to create Android and iOS apps simultaneously.
Challenges we Ran Into
Our biggest challenge was having two new members to Hackathons (out of 3). Our most experienced member spent most of his time on the hardware and both of the new members had no experience with Firebase, IOT, Hackathons, React, React-Native, or
Expo CLI. We spent the better part of 8 hours just trying to figure out what was going on and maintaining calm as we attempted navigating through an environment (literally) that was extraordiarily daunting. New language, new APIs, new to such extensive virtual collaboration, the list goes on... It's safe to say we walked away learning a lot.
Accomplishments that we're proud of
We were really happy to pull something together despite all the panic, confusion, and frenzy. All of us fell in love with the project that we were cultivating throughout the Hackathon, and it evolved into something that we felt could really help the world. It was interesting to see the interplay between hardware and software come to life in the form of an IOT App that could be downloaded across all platforms, and the journey helped us become better engineers as a whole.
What we Learned
Throughout our research on the project, we learned how serial communication works, the nuances of power distribution among sensors, and organization schematics for hardware systems. In software, we learned how to utilize libraries and APIs that can be found online, and we learned how to approach app development from the lens of a full-stack developer. React-Native was very difficult to grab ahold of, but well worth it in the end as our ability to use Frameworks to create broad-reaching applications greatly increased.
What is Next
The wonderful thing about this idea is that it can go in so many more directions. We could use networks to communicate with emergency services in the case of emergency situations faster than fire-detectors and faster than people. We have the ability to gamify our device by awarding those who stay in the cooking station more frequently. We could create a community system that allows everyone to share their wonderful recipes across the globe. We could add increased data analysis with machine learning that tells us more than the eye can meet about the sensor readings and its impact on food and safety. We could add calorie tracking, ingredient tracking, and grocery shopping into our app, and this is all just the tip of the iceberg lettuce.
We built our project using for React-Native.
In this directory,
npm test
to get our system working for both Android and IOS Apps.
Libraries Used:
react-native
react-native-firebase
react-native-svg-charts
@expo/vector-icons
react-navigation
expo
moment
react-native-stopwatch-timer
Built With
arduino
c++
esp8266
firebase
javascript
react-native
Try it out
github.com | thecookingbot | An IoT and data analytical approach to kitchen safety and accessible and hyper reproducible recipes! | ['Abinavraj Ganesh Sudhakar', 'Srikar Ganti', 'Chaturved Sumanth Lakkaraju'] | [] | ['arduino', 'c++', 'esp8266', 'firebase', 'javascript', 'react-native'] | 68 |
9,899 | https://devpost.com/software/feelies | Resources
Emotion Tracker Page
Live Graph
Home Page
Inspiration
We were inspired by the current world situation. People are isolated, alone, out of work and out of school, which can cause many emotions to run wild. We created feelies to help put what everyone feels into perspective to help them feel less alone.
What it does
The user logs their emotions, and feelies stores that data to feed a live chart. The site also links to many resources to help people deal with their emotions.
How I built it
The application was built with a flask framework. We used a webapp template from appseed.us called Pixel(by Themesberg).
Challenges We ran into
This was our first ever hack-a-thon, so everything seemed challenging. We attempted to divide and conquer, but we hit many bumps in the road. The backend was by far the hardest part because neither of us were extremely familiar with flask. We found the project not building several times, conflicting files in the repo, visuals not showing, you name it and we probably encountered it.
Accomplishments that I'm proud of
We are proud to have successfully built our application. It was a challenging project, but to have finished is very invigorating.
What We learned
We learned a lot about web development and time management, both of which are useful skills that we will carry out of this hack-a-thon and into our lives.
What's next for Feelies
We hope to further improve our site as well as participate in more hack-a-thons throughout the course of the summer.
Built With
bootstrap
css3
domain.com
flask
html5
json
linode
python
sass
Try it out
github.com
172.105.149.150 | Feelies | Feelies is a site that allows for people to log how they feel. It collects the information to show them that they are not alone in what they feel. | ['Nicholas Blackburn', 'Lauren Gayle'] | [] | ['bootstrap', 'css3', 'domain.com', 'flask', 'html5', 'json', 'linode', 'python', 'sass'] | 69 |
9,899 | https://devpost.com/software/hackathon-submission-85zdqh | Inspiration
Our inspiration for this was mostly to educate children(and in some cases adults) how to protect themselves during this pandemic.
What it does
It takes in your inputs and puts them into the code, pushing you onto different branches of the story.
How we built it
Using repl, we collaborated on a file that had all of our code together like a google doc, and built it by using several while loops and if statements.
Challenges we ran into
There was a lot of issues with the physical code and the actual writing of the script, as it was hard to agree on something and we were all pretty new to coding python.
Accomplishments that we're proud of
The final game! It was a learning process that taught all of us how to code python.
What I learned
Generally how to code! Most of us were new to coding Python, and it was really fun discovering all of the things that it can do, especially in a scenario like this.
What's next for Corona Escape
We want to try and add images to make it more visual.
Built With
python
Try it out
github.com
coronaescape.anisunsetskie.repl.run | Corona Escape | A text-based adventure game that teaches people how to stay safe during the pandemic! | ['Lizzie Manis', 'Min Yi Liu [Student]', 'Anisha Jaiswal', 'HanooStreet'] | [] | ['python'] | 70 |
9,899 | https://devpost.com/software/fantasy-finances | This is the logo we created for the website
We were inspired by the recently popular financial helper Cleo. Cleo is an AI that communicated with its users and helps them reduce spending and save money. We wanted to follow this concept, by creating a product that helps users financially while also being entertaining to interact with.
Fantasy Finances has a simulated stock market which users can buy and sell from using their given virtual currency. The simulated stock market pulls real prices to increase realism. Users can use profits to buy buildings in a town-building game.
The website is built using HTML, Javascript, and CSS. In order to pull the stock data, we used the Alpha Vantage API.
Our main challenge was implementing the API into our code, and extracting the correct stock data we needed.
We are proud of the multidimensional aspects of the website; we were able to integrate a virtual stock market into an engaging game. We feel that it is a pleasant user experience.
We learned how to implement APIs into our web applications, and we increased our skills in firebase as well as website development.
We hope to keep improving the website by upgrading the aesthetics as well as increasing the amount of educational resources available. A big goal is to also implement a user's real life financial data, using the plaid API. This will allow us to get user transactions, and help them save money in real life.
Built With
alphavantageapi
css
firebase
html5
javascript
Try it out
github.com | Fantasy Finances | Fantasy Finances is a website that teaches people to become financially educated and allows them to practice investing in a simulated environment. Profits can be used in a town-building game. | ['shivan19m Mukherjee', 'Akash Mahesh', 'Jay Prasad', 'Shaan Patel'] | [] | ['alphavantageapi', 'css', 'firebase', 'html5', 'javascript'] | 71 |
9,899 | https://devpost.com/software/tempest-he7ot3 | Welcome Screen
Preferences Page
Preferences Page
Results Page
Results Page
Inspiration
We decided to create an application that helps users plan when the best time to exercise is. As runners, we know how hard to can be to motivate yourself to workout and how easy it is to blame the weather for skipping a day. We decided to address this issue and help runners set a weather-based schedule. With coronavirus affecting everyone’s lives, we hope to bring routine to people’s schedules with workout structure.
What it does
When first booted up, the app shows a short description of what it’s used for. It then takes the user to the preference screen where they are prompted to set a range of temperatures they would like to work out in. This provides more personalized recommendations as different users have different climate tolerances. The preference page also asks for what time frame they would like to workout in. The schedule will then display the next 24 hours of availability. This is enough time for the user to check their phones in the morning and know when would be the most comfortable and convenient for them. It also displays the next seven days and whether or not they should take advantage of a sunny day or do an indoor workout instead.
How we built it
We ultimately went with an Android App built on Android Studio. Our app is built upon the DarkSky weather API, a service that provides weather data for any given location. We used a fusedLocationProvider from Android to get the devices data, which gave us coordinates for DarkSky. We parsed the JSON object returned by DarkSky with the OkHttp library. After successfully parsing, we were able to put varies data sets into arrays and display them the text and list views on our app. The XML and java files linked to show the user exactly what their schedule is!
Challenges we ran into
We originally wanted to make Tempest a website, but after several hours of attempts, we decided to play to our strengths and make an app instead. This was a little disappointing at first, since the Hackathon had just started, but in the end we were very happy with the final product. We also struggled a bit to get the virtual devices set up correctly to test our app, but we figured that out too.
Accomplishments that we're proud of
We're really proud of completing a fully functional app in just 36 hours. Most of us had never done a hackathon before, so we really tried to focus on the experience and participate as much as possible. Our app successfully does everything we planned on and we all learned a lot along the way.
What we learned
We learned a lot about communication and how crucial it is when working remotely with a deadline. We stayed on calls the majority of the weekend and were able to bounce ideas back and forth very effectively. We also learned a lot about app development, which none of us had much experience in before.
What's next for Tempest
The state of the world in 2020 has brought out the runner’s spirit in many Americans, not only for the health benefits, but also just as an excuse to get outside and enjoy the fresh air. With Tempest we can make sure that every runner gets their chance to enjoy the sunshine and evade the rainy days. But not only is the running community expanding, the tempest user base and features also has the potential to expand exponentially; we can add features to target all sorts of activities such as hiking and cycling, to best prepare Tempest users we can recommend the best type of clothing for their activities, and for the new wave of hygiene enthusiasts we can add a feature that tracks the least populated areas and coordinates the Tempest users to promote the ideal social distancing scenario. Once some popularity has been gained we can even link like minded Tempest users based on their activity preferences. With Tempest, we’ll make sure every runner outruns their rainy day and their stormy night.
Built With
android-studio
dark-sky
java
okhttp
Try it out
github.com | Tempest | We decided to make a weather app designed for runners, by runners. Users select preferences and can view a schedule to help them exercise during the week. We make sure every runner outruns the storm. | ['Evan Chang', 'Mar9318 Nunez', 'Lucas Balangero'] | [] | ['android-studio', 'dark-sky', 'java', 'okhttp'] | 72 |
9,899 | https://devpost.com/software/pop-up | Inspiration
When approaching recruiters, you have to print those long resumes that people barely glance through... Recruitment and building your professional network can be a challenging thing. There is a limit to the number of Linkedin requests you make in a day and people barely glance at your projects and your experience. Our idea is to create a new personalized AR Business Card that will be an add-on to your Linkedin profile; showcasing components such as videos of project work, links, AR headers (gifs), and much more!
What it does
There are two components to this hack:
There is a Web App for the AR component (where you can view the AR)
There is a Mobile App for creating and generating your personalized QR code.
Our app will add you to the our AR 'professional network' once the user signs up. It asks your personal information including:
Name, Description/Title, Quick Links (Github, LinkedIn, etc), Contact-Links (email, phone), Optional Headers (gifs which showcase who you are or a project), Optional Experience/Status
For those who don't have the mobile app, they can still view the user's info by scanning the QR code with their camera app.
How I built it
Our Mobile App uses Apache Cordova and the our forms were all built with Vue and React JS
The AR Web Component is built with AR.JS & React
Challenges I ran into
Watch the video! We describe all of them
Accomplishments that I'm proud of
Our app works well!
What's next for Pop Up
We want to turn this into an extension for Linkedin. LinkedIn has its own QR code component so creating an AR Business Card over lay on top of the LinkedIn QR code would be our next step. Also, we want to extend the scope of our current form so that it is more customizable.
Built With
ajax
apache
google-cloud
react
vue
Try it out
github.com
pop-up-ar.web.app
qrcodes-app.web.app | Pop Up | Business Cards are boring :( Create interactive AR/Static Webpage Business Cards that encompass all professional user information including: resume, LinkedIn, etc. | ['Aman Adhav', 'Roman Koval'] | [] | ['ajax', 'apache', 'google-cloud', 'react', 'vue'] | 73 |
9,899 | https://devpost.com/software/covid-19-global | Main Page, Search for Country Covid Data
Country Covid Data Page
Time Series on Covid Data
Inspiration:
Our idea was brought about by the need to address the concerns that we share about the ongoing Coronavirus pandemic. Particularly, we felt concerned with the spread of information. Among the innumerous online sources which cover Covid-19 material, it is sometimes difficult for the average person to find the exact information they want or know whether it is always coming from a trusted news source. With this in mind, we decided to create our own platform for Covid-19 information that would provide current and trustworthy facts and figures.
What it does:
Our project is a desktop application which begins by displaying the current Covid-19 values for the World. These include the total number of cases and deaths, as well as the new number of cases and deaths for the current day. The user then has the option to type into a search bar the name of the country they wish to receive more specific information on. The application would then direct to a new screen where the country flag is generated along with the current Covid-19 facts, health statistics, and demographic values for that country. Furthermore, the user can click on any on the daily Covid-19 values to be redirected to another screen which displays a graph of all recorded values for the country since the beginning of the pandemic. It is important to note that all information is taken directly from a regularly updated online database which gathers its information from health and government agencies such the European Centre for Disease Prevention and Control, UN, and World Bank.
How we built it
We used various types of libraries and processes to aid in our program's back-end. Requests was used extensively to retrieve data from the web in various formats such as json or html. The json was then dumped and formatted into a json file called
data.json
and was then parsed later using the Pandas library to create dataframes for our data. We created various functions such as downloader, updater, and loader functions that deal with our data. We also created functions that manage our temporary files by efficiently deciding when to retrieve files from the web. All of this was then turned into a module called
covid_data.py
that could be imported into the main python file that runs the program.
As for our front-end, we used PyQt5 as our framework for our GUI. We started of by templating and prototyping with the Qt Designer application so that we had a foundation to begin building our application upon. Once we were satisfied with a template, we converted it to python, and added more features to it later on. This code was then turned into a module called
design.py
that could be imported into the main python file that runs the program.
In our main python file, we created functions that could dynamically alter widgets in layouts as well as update values in them. We also used this file to bind buttons and labels to events and functions. The app's entry point starts from this python file.
Challenges we ran into
Matplotlib and pandas were libraries that seemed to us to have a major learning curve. We had challenges understanding how the library and its classes worked.
We had issues integrating matplotlib with pyqt5. It was difficult to embed the plots into the layouts within the qt GUI.
Accomplishments that we're proud of
We were able to make a project that was able to satisfy all of our initial goals.
Able to gather Covid-19 data from trustworthy sources
Allow user to search which country they want to receive data from
Created clear and organized user interface for users to view information
Able to generate graphs based on gathered data
What we learned
We were able to learn about many different data science libraries and also got a bit of experience on how to use parts of it.
This hackathon also gave us the opportunity to force ourselves to go outside of our comfort zone and tackle new things in python and data management.
What's next for Covid-19 Global
We would like to speed up our application significantly by getting rid of excess nested function callings in our code as well as redundant variables.
As well, in the future, we would like to package this application to be easily distributable to users of all major operating systems.
Built With
beautiful-soup
html
json
matplotlib
pandas
pycountries
pycountry
pyqt5
python
qss
qt
qtdesigner
requests
stylesheets
Try it out
github.com | Covid-19 Global | Desktop application that provides trustworthy statistics on Covid-19 from countries around the world. | ['Mahir Chowdhury', 'Ahmad Ali'] | [] | ['beautiful-soup', 'html', 'json', 'matplotlib', 'pandas', 'pycountries', 'pycountry', 'pyqt5', 'python', 'qss', 'qt', 'qtdesigner', 'requests', 'stylesheets'] | 74 |
9,899 | https://devpost.com/software/covid-19-world-count | Inspiration
The challenge to solve a problem inspired me. I was noticing that several people were facing challenges to track the carona virus (daily).Therefore,I created a program that could solve just that
What it does
Gives updates of the total number of carona viruses cases (through notifications)
How I built it
Python
Challenges I ran into
Less time to complete the project.
Accomplishments that I'm proud of
Completing my project (alone)
What I learned
Different syntax and functions of python
What's next for COVID-19 World Count
I hope to create an app in which the total number of carona virus cases would be recorded throughout the world. Furthermore, the app would also allow people to share an anonymous blog of how the virus affected them,
Built With
python | COVID-19 World Count | The main objective of this program is to track the carona virus cases throughout the world. | ['Raja Allmdar Tariq Ali'] | [] | ['python'] | 75 |
9,899 | https://devpost.com/software/platformer-tioej0 | Game Over
Gameplay
Starting Screen
Inspiration
I have always wanted to create an interactive mini-game that was both simple and inviting. The game we created reflected that. The player is a simple rectangle, while the enemies are circles.
What it does
To play this game, you must avoid the circles. Every ten seconds, more circles spawn. Once you touch the circle, you lose. The timer on the top right of the screen is your score.
How we built it
We built this game using Java and an IDE called Textpad. We coded everything from scratch and built a simple yet addicting game.
Challenges we ran into
We ran into many bugs during our development of the game. My team worked together and worked hard in order to solve as many bugs as possible during the given time frame. Overall, the game itself is smooth and easy to play.
Accomplishments that we're proud of
We are definitely proud of the fact that we built everything from scratch, without any form of game engine. The game's graphic are unique because they are so simple. Nowadays, graphics are extremely high-end. On the other hand, our game focuses on having a simple interface, which is much more user friendly.
What we learned
We learned how to use many different classes in the Java language, and we gained a much more deeper understanding of Java. Working together as a team was a valuable skill that we gained during the development of the game.
What's next for Platformer
Platformer is aimed to be simple, and so adding different levels with the same graphics should be our next step. We want the game to be diverse yet straightforward.
Built With
java
textpad
Try it out
github.com | Platformer | Boxes avoid boxes. | ['Abhi Patel'] | [] | ['java', 'textpad'] | 76 |
9,899 | https://devpost.com/software/blm | Front Page
Inspiration
The idea behind this project was to make a website that what support the Black Lives Matter movement, explaining what the movement is, providing resources so that people can be involved in the movement, and giving inspiration to people to make a difference.
How we built it
We basically used HTML and CSS to format the webpage, and used some JavaScript to add more effects to it
Challenges we faced
Both of us were pretty new to HTML and CSS, so it took us a while to really understand what was going on with our webpage, and how to get certain effects to work, and how to link certain things to eachother.
Accomplishments that we're proud of
We are proud of how our website turned out, even though we just started learning HTML and CSS
What's next for us
We are going to continue to make this website better by adding more information and designing it better!
Built With
css
html
javascript
Try it out
github.com | Stand With Us-Black Lives Matter | A website to support the Black Lives Matter movement | ['Aditi Tyagi', 'Vanita Sharma'] | [] | ['css', 'html', 'javascript'] | 77 |
9,899 | https://devpost.com/software/book-it-tfbs0n | Motivation
Sometimes, with a vague description, it is difficult to successfully find the title of an older book on Google. This is where Book.it! comes in. With a couple optimized algorithms and a database of over 15,000 titles that go way back into the 1800s, Book.it can find that one old classic you simply cannot find with your Google search.
Method
First, we load in our dataset with
pandas
. We follow by precomputing the frequency of each word in the plot summaries of each title and store the information in a list of dictionaries. We serialize this variable with
pickle
so that this precomputation (quadratic time complexity) is only run once.
For each query,
The terms are split into individual words.
For each title, the
tf-idf
value of each term is calculated.
A score is assigned to each title, being the sum of tf-idf values for each search term.
Titles are sorted by tf-idf values.
The top 20 results are taken. We calculate the
longest common subsequence
(an element in the sequence is a word) between the search terms and each title's summary. Using that metric, we finally reorder the top 20 results.
The time complexity for a single query is O(NlogM) where N is the total number of book titles and M is the sum of the number of words in the summary of each title.
Our choice to use LCS to rank the top 20 tf-idf results is to add grammatical structure to our ranking process. Although tf-idf by itself is a spectacular ranking algorithm, it is only limited to individual terms. LCS enables us to gauge how relevant the entire search string is (though it is obligatory to mention that LCS on its own, without the help of tf-idf to narrow down the results, is not a good algorithm to use for our purposes).
What's Next?
The next steps for Book.it! is to further optimize our search results. To do so, we can add a rating system for users to rate the accuracy of their searches and also support
A/B testing
. With A/B testing capability, we would be able to scientifically test which parameters work best for our search algorithms.
Built With
bottle
javascript
numpy
pandas
pickle
python
Try it out
htne20-book-it.herokuapp.com
github.com | Book.it! | Discover old classics that Google has forgotten. | ['Freeman Cheng', 'Ruyi Li'] | [] | ['bottle', 'javascript', 'numpy', 'pandas', 'pickle', 'python'] | 78 |
9,899 | https://devpost.com/software/retagged | Home screen
Upload an Image to get exact address and location
Explore around the place
Inspiration
We have all been in a situation where we see a beautiful picture of a place posted by someone who you follow on Social Media. We also would like to visit that place. Oh, wait! It doesn't have a location posted with it and it doesn't even have a caption? Also, you don't want to DM them specifically to ask this,
'coz we all have that inner shyness and fear of getting left on seen'
.
So, we decided to take a better step at helping humanity with this process. We have built an android app that recognizes the
exact location
of the place in the image you want to visit and recommends popular places near that location. Basically, we have your entire custom made trip planned.😏
What it does
Retagged is an efficient, custom-tailored travel app that combines Machine learning and Google Cloud API with Places API. Retagged helps you find (
Retagging
) those exact places from "those" images on Social media which does not have any info attached to it. We also suggest the most popular places visited by people around those places. This will help you save money by visiting places in a single trip or future trip planning.
How we built it
Started up by firing up the Android Studio. 🔥
Then some flow diagram drawings on the whiteboard. 🖼
Then comes the Google Cloud Engine with feature extraction.💪
Detecting location from the image. 👻
Geocoding and Reverse Geocoding here and there.♻
Extracting and displaying that location.⚓️
Making those custom recommendations based on the location retrieved 👀.
Parsing that JSON 🥺
Making UI elements and beautifying'em with Material Design.😎
Challenges we ran into
Filtering out images based on location and discarding all the other images with no prominent landmark was great challenge for us. We both learned great deals about Google Cloud Platform and making a custom recommendations based on the radius/distance of that geolocation.
Accomplishments that we're proud of
We have created a functional android app embedded with Machine Learning capabilities in a limited amount of time. AND IT WORKS ACCURATELY! 😳
Everyone tried new APIs and new tools for the first time on the project. 💪
What we learned
Getting an idea is a very important but successful execution and implementation should be the ultimate goal.
What's next for Retagged
Develop Retagged with more robust ML algorithms.
Develop a business models based on recommendations.
Developing more features for the better User Experience.
Built With
android
android-studio
foursquare
google-cloud
java
json
material-design
Try it out
github.com | Retagged | See More. Travel More. | ['Akshay Kulkarni'] | [] | ['android', 'android-studio', 'foursquare', 'google-cloud', 'java', 'json', 'material-design'] | 79 |
9,899 | https://devpost.com/software/thermomini | I started this project for the hackathons at the beginning of COVID-19, and have been slowly using every weekend since to learn the code that would allow this board to actually work, as well as improving the craft, this progress is slow and incremental, and thus, this project is more or less a placeholder demo.
Inspiration
How there's thermometers everywhere in China
What it does
Nothing. This is an Altium board that runs a thermometer in theory
How I built it
Working off of a previous board
Challenges I ran into
Team disappeared during Winhacks, no code
Accomplishments that I'm proud of
Having something?
What I learned
Erm... Hardware is only as good as the software that it runs on, and the people that make it. But also, Hardware is hard to do in a plague
What's next for Thermomini
Sleeeeep
Built With
altium | Thermomini | Small thermometers for COVID-19 | [] | [] | ['altium'] | 80 |
9,899 | https://devpost.com/software/mdeliver-xldzys | Rendered Map
Sign up for for users
Home
Inspiration
Because all of my Summer plans were canceled, I decided to come to Korea and meet my family. These past two weeks have been a new experience for me because I am quarantined at my house: I can't leave the house. Therefore, all the food I have eaten and all the things that I needed were ordered online and delivered to my house. However, I found this system very inefficient, where one person comes to my house with food and the other comes with tools or appliances. I thought that having one person deliver everything was a more efficient solution.
Also, another inspiration of this project was the current state of society during the Coronavirus. With this virus, many businesses and local shops have been decimated. While many companies were destroyed, a few others such as amazon and uber eats thrived. With Uber drivers and many other people unemployed, I think that MDeliver will not have a hard time finding a workforce, a network of delivery drivers. Therefore, MDeliver can potentially lower the unemployment rate. Also, MDeliver will cost less in various cases and will spur interest in various groups of people. Consider this case: a family is building a desk in their house and they are arguing about what they want for lunch. As they work, they figure that a screwdriver may be helpful in building the desk. MDeliver is perfect for cases such as this because MDeliver enables users to request multiple stops: a tools shop, a burger joint, and a Chinese restaurant.
What it does
This project is a submission for the HTNE Hackathon 2020. MDeliver hopes to make buying more convenient and fast. Rather than having delivery drivers deliver from one place at a time, MDeliver provides the option to deliver from places. Unlike other companies that offers the option to deliver from only one restaurant/food place, MDeliver aims to allow customers to order from multiple stores: not only for food, but for groceries, products, and others. More information is in the site linked.
How I built it
I used node.js for backend development and HTML javascript css for frontend development. The backend was created with express.js. For user authentication, I used a tool called bcrypt to create salt and generate passwords and socket.io for communication. I created the front end with html and styled it with some css and bootstrap. The app was made interactive with javascript mostly for the map portion of MDeliver. I used Microsoft Azure Maps for the map to place pins and create routes. All POST requests were handled with the help of express middleware and data about credentials (not passwords) was stored as cookies that were cleared after a new login request.
Challenges I ran into
I maybe should've used an online builder. I used bootstrap and created all elements by myself and this took a lot of time.
Google Maps API was giving me problems because I didn't want to pay money for it and was on the developer version. The directions API for google maps wasn't working because of my lack of subscription so I had to search for other solutions.
I luckily found that Microsoft Azure (paid by github student developer pack) had a maps API which was very good. However, I spent an hour trying to figure out why the cdn link wasn't getting the file and it turns out that South Korea(the country I am in right now) is one of two unsupported regions for Microsoft Azure Maps. -Therefore, I had to use a free trial of ghostvpn which was quite slow with every server available with the free version.
Accomplishments that I'm proud of
I am proud that I have a working version of my webapp. I didn't think I would finish in time because of all the challenges.
I am proud that I decided to finish this project because I was thinking of sleeping early.
I am happy that I learned many things along the way.
What I learned
This was my first time using something other than Google Maps API. This was a new experience and I enjoy Azure Maps. I also learned web designing along the way. I looked at courses on making webapps look aesthetically pleasing and I found it helpful and fun. Towards the end of the hackathon, I learned to use others as resources, such as the discord. It was a bit unfortunate knowing about the discord server and the help less than 24 hours before due date.
What's next for MDeliver
Creating a better page for POST success, an actual html page. (Same with other pages such as this)
Experimenting with it by getting clients locally and testing the process out. (Users and delivery persons)
Improving the website (both aesthetics and ease of usage)
Getting backers
For the map, allowing the navigation mode and creating the shortest route algorithm
Better estimations for prices
Creating a phone app
Built With
azure-maps
bcrypt
bootstrap
css3
ejs
express.js
javascript
node.js
npm
Try it out
github.com | MDeliver | The new Multi-Delivery Service. | ['Hyun Min Kim'] | [] | ['azure-maps', 'bcrypt', 'bootstrap', 'css3', 'ejs', 'express.js', 'javascript', 'node.js', 'npm'] | 81 |
9,899 | https://devpost.com/software/animetaro-6tuw1j | You Thought This Would be a Project Title. BUT IT WAS ME, DIO!
-----------------------------------------------------------------
Imagine This
----------------------------------------------------------------
You are browsing
MyAnimeList
for something new to watch and you come across something that looks junbi ok. You want to try it out to see if it's something you will like, but you don't want to go out of your way to look for a site to watch it on if it might be something that you don't like. If only there was an easy and seamless way to navigate to a website to instantly watch the anime you want to. Then
EKUSUPLOOSION
you find AnimeTaro.
The Chrome extension built with JavaScript to help you watch the anime you want to watch without having to search the web for a place to watch it.
Just click on the extension on any
MyAnimeList
anime page and get taken to a website to watch the anime in question! No more searching Google for a sketchy website to watch your favorite Japanese animations, AnimeTaro does it for you!
Step 1
Find an anime you want to watch (Such as
this
one)
Step 2
Click on the browser extension while on the page!
It's as easy as that!
Detailed instructions (installation and usage) on the GitHub
README.
This project was made for fun so please don't take anything too seriously... but it does work just saying...
Built With
chrome
css
firebase
html5
javascript
Try it out
github.com | AnimeTaro | Oh? You're Approaching This Project? | ['Vincent Li', 'Benjamin Saobuppha'] | [] | ['chrome', 'css', 'firebase', 'html5', 'javascript'] | 82 |
9,899 | https://devpost.com/software/decentralisedwifi | Schematic Diagram of Connection
Decentralised WiFi using Blockchain
Improve the existing WiFi Sharing technology according to the advancements in Blockchain Technology.
Problem Statment:
Create a
decentralised
authentication
mechanism for WiFi
Work as a peer to peer network.
Provide
pseudo-anonymity
to users.
A programmable network.
Create software for hotspot that controls the connection and authentication.
Users can connect to the WiFi through the Website that is loaded when connecting.
Selects the device randomly from connected devices.
## Advantages:
User doesn't need to install or maintain anything.
Compatible with the
existing
system
.
( Only some changes in the software; Mostly are invisible from the user side.)
Decentralised network.
Working:
The users are added to the blockchain once they connect to the WiFi (using their public address).
On connection, the user is verified by selecting a randomly connected device by WiFi Hotspot.
Registrations and verifications are treated as transactions and added to the blockchain.
To remove a user create a transaction of removing them.
Built With
html
javascript
Try it out
johndavistl.github.io
github.com | Decentralised WiFi | Decentralised WiFi using Blockchain | ['John Davis'] | [] | ['html', 'javascript'] | 83 |
9,899 | https://devpost.com/software/elixir-dmfj49 | Inspiration
Elixir is inspired by the typical planner everyone uses. We wanted to design Elixir to create a platform where people can easily plan out their activities while also being able to access other resources that make the scheduling experience a little more fun. The name “Elixir” is inspired by its definition: a magical or medicinal potion. We wanted the user to be able to create their own magical potion that acts as a guide to be more productive and improve their daily life.
What it does
Elixir is a website designed to allow users to input a to-do list with the tasks that they want to complete throughout the day. As the tasks get completed, the activity is moved to a completed section and if the user wants to delete the task, they can simply press the trash icon. However, we wanted to add additional options that elevate the planning experience. After clicking the advice icon, the user will be taken to a page asking about their feelings for the day. The user will then be taken to a page with information about their feelings and gain advice on what to do. In addition, we created a daily challenge option for the users so when the wheel is spun, a random activity will appear on the screen and give the user a goal to accomplish for the day.
How we built it
To create the graphic designs and background on each page, we used Canva and the templates provided to help guide us. Then, we used Figma to determine the layout of each page and connected the pages to one another so that when the user clicks on a certain component of the page, it would take them to the right location. With Figma, we were able to export all of the final designs along with the CSS code to Visual Studio Code (VSC) where we programmed everything together in HTML and javascript. In VSC, we were able to call the CSS code in HTML to access the layout and properties of each component in order to have our website designed the way we liked it. Also, after many attempts to build the to-do list and spinner, we had to resort to using sources found online and will give credit to them under “challenges we ran into”.
Challenges we ran into
It was our first time programming in HTML and CSS and we have never built a website from scratch before so we faced many challenges throughout the learning experience.
Our first challenge was navigating our way through Figma and Visual Studio Code as we watched many tutorials to learn the different tools we could utilize. We could not figure out how to make the images appear on the screen and also how to make buttons so that the pages would be linked to one another. This is a helpful resource to add buttons:
https://www.w3schools.com/tags/tag_button.asp
Also, these two YouTube videos by Web Dev Simplified were our favorite videos that really helped guide us throughout the process when designing our own website:
https://www.youtube.com/watch?v=RZ-Oe4_Ew7g&t=151s
and
https://www.youtube.com/watch?v=FK4YusHIIj0
Another challenge we faced was figuring out how to get the user input in order to create the to-do list and also figure out how to create a spinner for the daily challenge. We looked up resources to help guide us in the process, but errors continued to appear and it became difficult to design it in a way where it looked aesthetically pleasing. For instance, with the to-do list, we were able to get the user to type in the box by programming in javascript, but we were unable to continue through with the if-statement so when the user pressed the “enter” key the activity would be listed. We ultimately had to resort to following the online tutorial and example code and would like to give them credit for it. Here are the links we used and modified to fit our ideas for Elixir:
https://www.youtube.com/watch?v=h7gZY3_3Dqs
and
https://code.sololearn.com/WD8GprR5hozY/#html
Lastly, we faced the problem of not being able to have the layout of the website to fit the screens of all devices. We researched the problem and looked at the recommendations from other people and tried to implement some into our own program. However, none of it seemed to work and we continued to get errors about the sizing. Although we could not solve the problem during the time of the hackathon, we hope to be able to figure out how to do this for future projects.
Accomplishments that we're proud of
We are proud of learning so much through this project specifically, HTML, CSS, Figma, Visual Studio Code, and the overall process of designing and programming a website from scratch. We are proud that we are able to have a working website to look back on in the future and see the progress that we have made. We are glad that we were able to turn this hackathon experience into a learning experience and create a project using tools we have never used before.
What we learned
We learned how to build a website through HTML, CSS, Figma, and Visual Studio Code. Within that, we were able to learn the ways of connecting pages to one another through the button tool and also write in javascript for the user input.
What's next for Elixir
We originally started creating Elixir hoping to design a better calendar for everyone to use. We tried to use Google Calendar and Google Sign-In so the information would be linked to the user’s Google account. Although we could not complete this during the hackathon, we hope that in the future Elixir will become a platform where people could easily access a calendar and schedule calls for certain times. We also hope that we can program it so the information will be saved somewhere for the user to always access it. It would also be nice to be able to create a page that would generate a daily inspiration quote for the user or any other fun ideas they could implement into their life.
Built With
canva
css
figma
github
html
javascript
visualstudiocode
Try it out
github.com
sanjum937.github.io | Elixir | A website designed to improve your daily activities through inputting a to-do list and receiving daily challenges! | ['Sabrina Chang', 'Sanjum Sahni'] | [] | ['canva', 'css', 'figma', 'github', 'html', 'javascript', 'visualstudiocode'] | 84 |
9,899 | https://devpost.com/software/studbud-htne | Main page of StudBud
Join an existing StudBud Group
Create your own studbud group
StudBud-HTNE
Hack the North East Hackathon Project
Most Viable Startup Track
www.studbud-htne.tech
Study Groups for Students by Students
For students in high school and college who would benefit from a group study session, but want the flexibility to choose your time, place, and frequency, then StudBud is for you! StudBud is a web application that allows students to easily connect with other students to form a study session on demand. As a student, I know that often it would be super helpful to meet up with students in my class before an exam or before homework is due, but to have the flexibility I need that fits my need and desire. To overcome this issue, StudBud provides a unique opportunity on the market, as there is no strong competitor aside from group chats. Yet group chats fail when you're placed into a class with students you do not know.
From the beginning of this hackathon, I chose to use technology that was completely knew to me - to challenge myself and explore new powerful services. For this project, I wanted to make it possible to scale to any number of users provide a enjoyable user experience, so I learned how to make a serverless backend through AWS and create a frontend through Vue Js.
Running StudBud
StudBud is broken up into two halves - front and back end.
Frontend
The frontend of the application was developed using Vue Js with Axios to fetch calls to the API. To serve the application, any package manager for Node.Js will work.
Deploy Frontend Locally
Install dependencies with
npm install
Run locally with
npm run serve
Dependencies:
"@mdi/font": "^3.6.95",
"axios": "^0.19.2",
"core-js": "^3.6.5",
"roboto-fontface": "*",
"vue": "^2.6.11",
"vue-cli-plugin-s3-deploy": "^4.0.0-rc3",
"vue-ctk-date-time-picker": "^2.4.0",
"vue-router": "^3.2.0",
"vuelidate": "^0.7.5",
"vuetify": "^2.2.11"
Backend
The backend of the application was developed all serverless. As execution and demand were key components to StudBud's design process, it was decided to use AWS for all backend hosting. The tools used were Amazon Lambda, Amazon Api Gateway, and Amazon DynamoDB. By going serverless, StudBud has the huge potential right from the start to instantly scale to any number of users without needing to manage and provision it's own backend servers. This gives StudBud immediate room for growth to expand to mobile devices and focus on user experience rather than maintenance.
Amazon Lambda
The Lambda node.js files used for developing the functions are in the backend directory.
Authors
Alex Jalkanen
Built With
html
javascript
vue
Try it out
www.studbud-htne.tech | StudBud | StudBud is a web application that allows students to easily connect with other students to form a study session on demand. | ['Alex Jalkanen'] | [] | ['html', 'javascript', 'vue'] | 85 |
9,899 | https://devpost.com/software/paimon | deleted
Built With
javascript
Try it out
jonin.gq | paimon - deleted | The bot has been shut down. | [] | [] | ['javascript'] | 86 |
9,899 | https://devpost.com/software/protest-app | Inspiration
2020 has been a tumultuous year filled with raw and authentic emotions that will forever resonate throughout the history of our kind. One such monumental movement is that of Black Lives Matter and demanding political equality for those who have been silenced for centuries. Protests.Me is a web based application that allows users to locate protests occurring near them and submit protest forms for the public to attend. It allows users to see how many participants are intending to attend an event, the date, time, location, and important information such as items to bring.
What it does
Protests.Me allows users to create and participate in protests occurring within their vicinity. Upon accessing the web application the user sees a feed of protest locations pinned around their area. They can choose the pinpoints on the map to see additional information (date, time, number of participants), as well as an option to attend. Logging in allows the user to save their joined protests, and get reminders for upcoming protests. It also enables them to assemble their own protests and spread relevant information about it.
How we built it
This project’s frontend was built primarily in HTML, CSS, and BootStrap. The backend was built and deployed on Firebase in order to enhance features like connectivity to other social media platforms and authentication. In order to connect our front and back ends, we used node.js. The protests that were created were stored on Firebase.
Challenges we ran into
As a team, we ran into challenges and difficult decisions when discussing features that we wanted to incorporate within our application. We all had a diverse amount of ideas, all which are necessary for an application, but had to prioritize those that we felt were critical to maintaining the integrity of the application given the time restrictions. Additionally, we found ourselves facing difficulty when it came time to implement data transfer and display protests on our frontend from the data stored on Firebase.
Accomplishments that we're proud of
Learning and using a new devops/backend service Firebase to bring together our end product is the biggest accomplishment we are proud of. As mentioned previously, none of us had experience with Firebase, so we are proud of the functionality we were able to implement through it. We are also proud of our two teammates who pushed through and learned some HTML/CSS/git so they could help in creating the project. Although they had zero experience with web-development prior to this, they were persistent and helped create our UI.
What we learned
As a team, we learned collectively how to communicate and harness each other’s strengths in order to create an application that we were all satisfied by. We coordinated and discussed throughout the days in order to ensure that all of our teammates were up to speed on what was going on. Additionally, we all learned a lot about Firebase as none of us had experience with it prior. Firebase includes a multitude of features that we were unaware of going into it, however, they were very simple to implement.
What's next for Protests.Me
In the future, we will look to expand Protests.Me so it is accessible on mobile platforms such as iOS and Android as well. This would allow for a greater reach and more users to take advantage of the services it provides. We will also be adding a live feed of protests from participating protestors and the ability to attach photos to previous protests. We will also implement a Public Service Announcement functionality so that individuals are able to send messages concerning protests for others to be aware of.
Built With
bootstrap
css
firebase
html
javascript
node.js
Try it out
github.com
protestsdotme.web.app | Protests.Me | Protest Visibility to Spread Awareness | ['Chethin Manage', 'Minh Nguyen', 'Ashmita Rajkumar', 'ochoav Ochoa'] | [] | ['bootstrap', 'css', 'firebase', 'html', 'javascript', 'node.js'] | 87 |
9,899 | https://devpost.com/software/hackthenortheast-6divsm | Fake News Linear SVC Model
Hack the Northeast Hackathon Project --
Predicting whether an article is fake news or trustworthy based on its title.
Fake news has been in the news a lot since 2016. However, in the last few weeks, social media companies Twitter and Facebook have grappled with the idea of labeling false information on their sites. Twitter has started to flag certain tweets, but they are currently based on hard text within the tweet, or its done by hand. Facebook has said it will not attempt to label false information.
This inspired us to create a machine learning model to predict whether an article is fake news based on its title. This would allow Twitter or Facebook to check the title of an article being shared and determine whether to flag the tweet or not. This algorithm would have huge applications as the 2020 election gears up.
In creating our model, we cleaned and separated words from the title using the nltk library. Then, we used Term Frequency * Inverse Document Frequency to get floating-point values for the most common 300 words per title. Finally, we used sci-kit learn to create a linear support vector classifier model that predicts whether an article is fake news based on the title with 98% precision.
Built With
jupyter-notebook
nltk
pandas
python
scikit-learn
Try it out
github.com | Fake News Linear SVC Model | Using a linear support vector classifier model to predict whether an article is fake news based solely off of its title | ['Seth Keim', 'Vipul Periwal', 'Amra Mendoza'] | [] | ['jupyter-notebook', 'nltk', 'pandas', 'python', 'scikit-learn'] | 88 |
9,899 | https://devpost.com/software/music-player-fjo4z3 | Blynk
Arduino
Inspiration
When I learned about the arduino piezo alarm, I recognized the wide applications it could have. I could play
my own songs
! So, I started looking at current arduino programs that play songs based on frequency, but I saw that the notes and frequencies for the songs done manually for a specific song. So I decided to make a way to play any song you want rather than converting all the notes in a song to fundamental frequencies manually.
What it does
I built a javascript program to convert any musicxml file that plays one note at a time to the exact arduino code needed to play the song. In converts the notes in the songs (A, B, C, ..) into the exact fundamental frequencies needed to play the song on the piezo alarm (basically an arduino speaker). So all you need to do is copy-paste the arduino code into the arduino application and run the code, and hear your song on the piezo alarm. I also enabled blync (an iot app) for this program, so you can stop or play the song remotely with your blynk app. So, essentially, I created a bluetooth speaker using just a piezo alarm and an arduino.
How I built it
I used an ArduinoMKR1010 (a wifi-enabled arduino), a battery, and a piezo alarm for the hardware. I used javascript for the software, and the javascript outputs the exact arduino code you need to play any song you want in the terminal.
Challenges I ran into
Figuring out how to parse MusicXML files was very difficult since there is just so much text, so I converted it to JSON and used a JSON editor to find how to get the durations, octets, and letter for every note in the song. Then, I used that information to produce an array of the durations of every note and an array of the fundamental frequencies of every note.
Accomplishments that I'm proud of
I made my own bluetooth speaker, and at a much cheaper cost because of the cheap components!
What I learned
I learned how to use blynk to control Arduinos remotely more easily, and it was fun sliding the slider in the blync app up and down to play and stop the song!
What's next for Music Player
I'm thinking of making the front end of the program using react so the user can just copy paste music xml into a webpage to get their arduino code instead of doing it in the code itself. I could also have a store of previous songs
Built With
arduino
blynk
c++
javascript
Try it out
github.com | Music Player | Use an arduino to play any song you want! | ['ram potham'] | [] | ['arduino', 'blynk', 'c++', 'javascript'] | 89 |
9,899 | https://devpost.com/software/northeasthack | Main Page
Grammar exercise
Division exercise
Addition exercise
Name the image exercise
Match the image exercise
Inspiration
Millions of people worldwide suffer from strokes. Those people who have experienced one need a deep and lengthy process of rehabilitation. One of the difficulties that the person might experience is permanent brain damage, including memory and visual problems, and obstacles in decision-making. The brain often compensates for the damage caused by stroke. Some of the brain cells that do not die may resume functioning. Sometimes, one region of the brain takes over for a part damaged by the stroke. Stroke survivors have experienced remarkable and unanticipated recoveries that cannot be explained. The recovery must be accomplished in a way that preserves dignity while motivating the survivor to relearn basic skills the stroke may have taken away. Our website aims to provide patients who have recently undergone a stroke with a tool that will aid in brain recovery. The website implements four types of exercises that help the patient regain cognitive abilities, logical thinking, and speech understanding. Studies show that those interactive brain exercises show effectiveness preclinically as an intervention for stroke. Stroke survivors often find that once-simple tasks around the house become extremely difficult or impossible. The tasks and techniques on our website are available to help people retain their brain function safely and clearly.
What it does
Our Project Brain Recovery is a web app made to help people who recently had a stroke or any other illness and got temporary damage to their brains. In the USA only 795,000 people have a stroke every year, from them about 610000 are first or new strokes. Our web app has a few exercises which help to recover brain functionality. Doing small exercises like arithmetics or matching the pictures helps patients to start differentiating between objects and helps them recover and get back to their daily life easier. One of the developers had a similar experience as his father also had a stroke and to help him recover doctors made him do similar exercises. Moreover, from their experience we can tell that this is very effective, one difference is that our web app lets people do these exercises everywhere if they have the internet connection and also tells them if they picked the correct answer. Finally, they can choose how many tasks of one exercise type they want to do.
How we built it
We used various technologies in the production of our project such as the Vaadin framework, Spring framework, Heroku Cloud. The main programming language for that was Java and we also used a bit of CSS to style our Vaadin UI components.
What's next for BrainRecovery
We are planning to create a full-fledged tool for medical institutions by connecting our app to the database and enabling users to add more custom tasks. As for the patients, the website will have the statistics and progress tab, which will help both them and doctors to monitor how their rehabilitation goes.
Built With
heroku
java
spring
vaadin
Try it out
brainrecovery.herokuapp.com
github.com | BrainRecovery | A web application that helps post-stroke brain rehabilitation. | ['CSAjchan Mamedov', 'Pavel Petrukhin', 'Cian Jinks', 'Shohinabonu Shamshodova'] | [] | ['heroku', 'java', 'spring', 'vaadin'] | 90 |
9,899 | https://devpost.com/software/article-summarizer-0s2n53 | Index Page
Sample URL
Output, A perfect summary that reduces the amount of reading significantly
Inspiration
I had little time to look through the news article and even then I was not able to finish reading the news. This inspired me to create a fast way to read the lengthy articles in a short period of time in order to have time to finish reading the news.
What it does
My project helps to summarize web articles to reduce time that is spent reading the news. The must paste the link of the article and hit submit. Then the website and/or app will shorten the article, anywhere form a sentence to a paragraph, allowing the user gain the same amount of information while utilizing less time which then can be used for something more productive.
How I built it
This project was built using python natural language toolkit and personally scraping and implementation algorithms using beautiful soup. The whole interface was written in a combination of HTML/CSS/JS. Additionally we were able to implement mobile version by using heroku online hosting
Android Download Link:
https://s3.amazonaws.com/gonativeio/static/5edabab146111c28c199603a/app-release.apk
Challenges I ran into
Switching from web flask to website, coding for texture, and using user interface. I tried to create a button that would allow the user to go back to the summarizer page, however I was only able make the website and app, but the user can refresh the app or go back to the summarizer page using the back button.
Accomplishments that I'm proud of
Successfully changing a web flask into a website and app, being able to create appealing appearance, highlighting as the user reads, and using user interface.
What I learned
I learned how to convert a web flask to a website, code for texture in python, and create user interface.
What's next for Article Summarizer
Being able to summarize multiple articles with having to refresh or go back, allowing user to choose summary length, and diversity in language options.
Built With
csi
html
javascript
python
Try it out
nlp-article.herokuapp.com
www.theverge.com
github.com | NewsPoint | A short sweet summary of long articles! | ['Neha Singhal', 'Aditya Singhal'] | [] | ['csi', 'html', 'javascript', 'python'] | 91 |
9,899 | https://devpost.com/software/pills-on-wheels | Logo
Home Screen
Login Screen
Register Screen
Client Homepage
View Account
View Prescriptions
Add Prescription
Order Prescription
Order Confirmation
Driver Homepage
Driver Choose Delivery
Driver Delivery Confirmation
Inspiration
This app was inspired by our entrepreneurial spirit. We decided to create an app that would improve the health and safety of people in North America, while also being a viable startup. All three developers have passion for software development and entrepreneurship, making this the an extremely well-designed app.
What it does
Secure login system, with personal accounts for drivers and customers
Customers can add prescriptions, which can then be viewed or ordered from a pharmacy
Upon ordering of prescription from customer, driver's available deliveries updates in real time to allow driver to accept any delivery in their area in their area
Smooth in-app experience, easy-to-use UI for both customer and driver
Secure storage of backend data, fetch requests used to push and pull data as needed
Verification on every form, cannot enter empty fields or invalid emails/mobile number
How I built it
This app was built in React Native using the Expo Client. This app is designed for Android users, and will be available on the Google Play Store in the coming days. The backend is built in the Python Flask framework and is hosted on Heroku.
Challenges I ran into
For all three developers, React Native was not our main language. Learning React Native throughout the hack was extremely challenging, yet super rewarding to see the amazing app we ended up building.
Accomplishments that I'm proud of
We are extremely proud of the final product we developed, given our React Native prior skills. We are also extremely proud to have taken an idea from idealization to a minimum viable product for our startup.
What I learned
We significantly improved our React Native skills and learned a lot about it. We also learned the challenge it is to make a well-designed application in the short timeframe.
What's next for Pills on Wheels
Next for Pills on Wheels is to take this minimum viable product and turn it into an actual startup. We will continue to develop our application while also seeking approval from the FDA and MHTA to allow us to use our app. We feel we have a well thought-out business plan (see attached business plan) that can lead us to success as a viable startup.
Built With
expo.io
flask
react-native
Try it out
github.com
expo.io | Pills on Wheels | A mobile app designed to help users organize and view their prescriptions from their phone, as well as order prescriptions straight to their home in contac-tless delivery. | ['Areez Visram', 'Neil Lobo', 'Reezan Visram'] | [] | ['expo.io', 'flask', 'react-native'] | 92 |
9,899 | https://devpost.com/software/lungviewer-e2foca | VR
AR
Inspiration
I was checking the impact of smoking and the health effects of Cigarette Smoking and the kind of impact it may have on Covid-19. I wanted to do something to prevent smokers from smoking. Perhaps something which gives viewers/users a jolt visually which motivates them. I thought of creating this app in which the users can see visually what kind of impact smoking is causing to their lungs and the damage smoking has caused to their lungs already which they can see visually.
What it does
LungViewer allows users to see how their lungs look like and the damage smoking has caused to their lungs dynamically. Users can see how smoking has progressively damaged their lungs over the years. Users are given an option to choose between AR mode and VR mode. In AR mode the user can hover the device on an image to launch the model. Users can slide the parameters like their age, how much they smoke, and years they have smoked. They can visually see the progressive damage smoking has caused to their lungs over time using the latest AR/VR technology. Not only this, but the app also educates the user on the differences between a healthy lung and a damaged lung, myths and facts, images, and more. Users can configure and pair their own Google Cardboard with the app using the settings gear icon in the VR mode of the app. The beauty of our app is that it is native and it works with both IOS and Android.
How I built it
Using Vuforia’s image targeting system and the Google Cardboard SDK, the app grabs an image of the lung viewer logo, which is uploaded to Vuforia’s database. It is then retrieved through Unity and used to show different stages of damaged lungs displayed on top of the image showing the damage based upon the parameters chosen by the user. Using Google Cardboard’s SDK this app carries the same functionality of the AR mode but through virtual reality, so users can see the lungs in more detail visually.
Challenges I ran into
It was difficult to tweak the model and the assets to fit with the project. I had challenges importing the Google Cardboard SDK into Unity using the prefabs that google provides was a bit challenging too.
Accomplishments that I'm proud of
My first experience using 3d models and Vuforia in Unity is a success. I usually use Swift to create apps but my first experience with making app with Unity.
What I learned
I learned how to use Vuforia through Unity and how to project a model on top of an image using Vuforia’s image targeting feature. I also learned how to toggle the VR Mode through the google cardboard SDK.
What's next for LungViewer
By choosing an image the targeting feature gives flexibility and more possibilities to raise awareness and educate users on other parts of human anatomy. In the future, I can add more information and textures for the lungs, animations, and rotating and scaling the lungs on the user’s command.
Built With
c#
echoar
unity
vuforia
Try it out
github.com | LungViewer | Raising awareness to smokers about their lungs through AR and VR | ['Krish Malik'] | [] | ['c#', 'echoar', 'unity', 'vuforia'] | 93 |
9,899 | https://devpost.com/software/allowance-amkn9z | Inspiration
Finances are hard. As kids, money seems so powerful. You get a crisp $20 bill and you feel like you're on the top of the world. Why does that feeling vanish. What can be accomplished to make finances feel less like a burden and more like a competition. We set out to solve this. We wanted it to be more than just budgeting. But empowering you to take control of your money and accomplish those goals.
What it does
Allowance lets the user plan financial decisions, save towards their goals, and win at financial empowerment. It automatically tracks the player’s wealth and credit score, meaning the only thing they have to do to become financially empowered is play. To play, players choose a financial goal like buying a house or starting a family, and can see professionally recommended preparedness steps to help them attain that goal. Players compete against each other, turning a boring task into a fun competition.
How I built it
Because allowance is meant to feel like a game, (and for the sake of time) we chose unity as our primary technology. This allowed us to avoid tons of boilerplate code and instead focus on implementing our idea. It also doesn’t constrain us to a particular platform.
Challenges I ran into
Unity is tough to work with for source control managers. So working on the project with teammates was a challenge doing it remotely. We solved this by minimizing the amount of changes done that could conflict during each push.
Accomplishments that I'm proud of
Our team is proud of the application. We took an idea, and turned it into a fully functional proof of concept. The prototype we built does not have everything that we envisioned, simply due to the time restrictions. But we are proud of the working application that shows the workflow.
What I learned
We learned that Unity is overkill for this application. If we could do it again we definitely would use Unity however. Unity is so powerful because you don't need tons of code to make simple things work and look good. Because of time constrictions this was a huge factor in our decision.
What's next for Allowance
Being a startup, we would love to see this app go somewhere. Security is always a risk when considering bank account information, so the first thing would be to create a secure application that can utilize bank APIs.
Built With
c#
unity
Try it out
github.com | Allowance | With allowance, financial management is as simple as Plan. Save. Win. | ['Luke Fleck', 'Hunter DeMeyer', 'Jordan Oberg'] | [] | ['c#', 'unity'] | 94 |
9,899 | https://devpost.com/software/watera | Inspiration
A lack of clean and available water threatens one of our most basic human rights, and we hoped to address both the issue of not knowing if your water is clean and effectively communicating the water quality of a community to water suppliers.
What it does
Watera combines current water quality information from hardware devices in a community with historical data to help households and communities plan for extreme weather events, provides resources relevant to your current water conditions to help with education and proactive water management efforts. Thanks to our hardware solution, even when severe weather events have made a connection to the internet overly difficult, community members still have the information they need about their water quality.
How I built it
The app was built with Flutter, the data is stored with MongoDB Atlas, an Arduino collects water quality data and sends it to the database via python.
Challenges I ran into
Connecting the hardware to MongoDB, collaborating on the Flutter app remotely.
Determining accurate measurements for the resistivity of water (had to make assumptions to make calculating the integral easier)
Accomplishments that I'm proud of
Getting the hardware to collect water quality data, we didn't have water quality sensors, so we designed and built a cheaper alternative to test water contamination.
What I learned
First time using MongoDB, learned how to connect to and retrieve data from the database.
Designing sensors to give
accurate
measurements is hard.
What's next for Watera
Building a web application to make it easier to access the data from anywhere.
Creating a network of these devices in a community that can communicate with each other in real-time to produce insights for both water suppliers and consumers.
Built With
arduino
c
flutter
mongodb
python
Try it out
github.com | Watera | An app and hardware solution that helps households and communities prepare and cope with severe drought or storm events. | ['Alex Yu', 'Dhrumil Patel'] | [] | ['arduino', 'c', 'flutter', 'mongodb', 'python'] | 95 |
9,899 | https://devpost.com/software/covid19-tracker-kit | Inspiration
During this corona outbreak, we are not able to identify whether a person roaming outside has corona or not. So, we got inspired from this and decided to create a kit where the people around can also confirm that the person has no symptoms of Corona or not.
What it does
It shows whether the person has corona or not, it diaplays the status of the person on a watch type of a band. Which he can show whenever he wants to travel, or go to a mall or go to a cinema hall.
How I built it
We built it using the different IC's and a NodeMCU platform to connect them. also we used some machine learning algorithms to detect whether the person has corona or not. the final product is a watch kind of a thing, which is easy to wear and fashionable.
Challenges I ran into
The challenge here was how do we find that the person has corona or not. So we went onto the ground zero and checked the symptoms of it. And dealt with it to design the product
Accomplishments that I'm proud of
I am proud to get the accuracy we dreamt of. Also, collecting the information from IC's and sending it for manipulations was a very tough job which we were successful to crack it.
What I learned
Here, we learned the machine learning and data science tricks and manipulations.
What's next for COVID-19 TRACKER KIT
The next part would be to bring it and implement on real life.
Built With
c
python
Try it out
github.com | COVID-19 TRACKER KIT | Tracking COVID with accuracy | ['PRIYAM266', 'Nihar Sanda', 'Satyam Thakur', 'Uddesh Kshirsagar'] | [] | ['c', 'python'] | 96 |
9,899 | https://devpost.com/software/honey-heist | Inspiration
For the design I have had the thought of this kind of game for awhile and its only a small fraction of what I wanted, truthfully this would just be a mini game in the real thing. Just didn't have all the time I wanted though.
What it does
Just a simple infinite runner to keep you busy, though it is harder then your normal runners which would definitely bother some people to either rage or become determined to do the best.
How I built it
With Unity, C#, Maya, and clip studio.
Challenges I ran into
I was having issues getting some parts of the code to read correctly put I solved most of my issues thanks to static variables
Accomplishments that I'm proud of
I do like how it looks and the way the combs spawn so you aren't forced to die since they are random.
What I learned
I was just learning/brushing up on some simple coding skills
What's next for Honey Heist
I would probably work on the actual game and even make more minigames. Right now though on its own I would like to refine the controls and make it for a mobile platform so that it can be used on its own. I would even like to add more themes and maybe levels of difficulty.
Built With
c#
itch.io
unity
Try it out
github.com
brianafigueroa.itch.io | Honey Heist | Simple infinite runner minigame, there rest would be more of a puzzle game. However this part on its own can be highly addictive as the difficulty level makes those who are determined to win come back | ['Briana Figueroa'] | [] | ['c#', 'itch.io', 'unity'] | 97 |
9,899 | https://devpost.com/software/viralcheck-social-media-app | Web app
Built With
python | ViralCheck | Web app | ['Jeremy Nguyen', 'Gideon Grinberg', 'Ritvik Irigireddy', 'Nand Vinchhi'] | [] | ['python'] | 98 |
9,899 | https://devpost.com/software/valueator | Home Page
Home Page
Registration page
Calculator page income
Calculator page expenses
Visual representation of your expenses
Inspiration
Every year there are new digital subscription services releasing providing various new things. Paying for these services such as Hulu, Netflix, Disney+ and more can add up in a hurry and who has the time to actually use all those services each month! We as a team wanted to create our own finance tracker that focused on the monthly subscription base services, and seeing whether or not it is really worth it for us to keep.
What it does
After putting in your monthly income, and adding your expenses for a month, you will be shown a chart that breaks down the different categories of how you spend your money. Our application will also show how much you are paying hourly for the online subscription based services.
How we built it
Valueator was built using the django framework, bootstrap 4, and along with some javascript and jquery.
Challenges we ran into
Our biggest challenge was trying to come up with an idea that would separate our product from the rest. This is how we came to the conclusion of focusing on the subscription based services.
Accomplishments that we're proud of
Django is a somewhat new experience for our team and we feel that we made a pretty decent application in the 36 hours we had to work on it.
What we learned
Our group learned how to product manage by assigning tasks to one another so that we wouldn't conflict with what the other person is working on at the time and kept strong communication throughout development.
What's next for Valueator
There are a lot of improvements that can be made to Valueator such as suggesting alternatives to certain expenses or possibly having the ability to compare your expenses to others in the same income bracket. A possible feature is where users can link their screen time data from their phones can be used to calculate an accurate hourly cost for digital subscriptions if an phone app version of this were to be created. The room for improvement is endless.
Built With
django
javascript
jquery
python
Try it out
github.com | Valueator | Subscriptions pile up without you knowing. Manage your subscriptions and take control of your financial life today. | ['Ken Tung', 'Trey Sumida'] | [] | ['django', 'javascript', 'jquery', 'python'] | 99 |
9,899 | https://devpost.com/software/crowdata | Make a listing requesting specific data for you machine or deep learning needs.
Answer other people's listings by providing data you have access to.
Check the listings you made and download the data you received.
Inspiration
I am very interested in deep and machine learning. I play around with pytorch, keras, and scikit learn often. However,
finding and acquiring the specific data that I need is often hard. The data available on the internet is too general or does not suit my purposes. I created crowdata to solve this issue.
What it does
Crowdata is a platform where users can request data and other people who have access to that data can submit it to earn monetary rewards. For example, say you need microscopic images of a cell when it was treated with a certain chemical to train a convolutional neural network. However, you don't work in a lab and specific images of these cells aren't available online. So, you go to crowdata and make a listing for the specific cell images you need. Then, another crowdata user who happens to work at a lab with access to these cells sees your listing. This user goes to their labs, takes the images you requested, and submits them to you through the crowdata platform. They are paid the price that you set, and you receive the images you need to train your deep learning model. This way, getting specific images for machine or deep learning is no longer an issue.
How we built
Crowdata was built using the MERN stack. MongoDB was used to store user info, listings for data, and the data itself in the form of base 64 images. Express was used to send http requests to the MongoDB database. POST requests were used to add new users, update users' balances, add new listings, and add data to the listings asking for it. GET requests were used to retrieve listings to display to the user. React JS as well as react-bootstrap was used to make the front end user interface where people could make listings and answer other peoples' by supplying data. Google OAuth was used for the login. Finally, node.js was runtime of the backend. To try it out for yourself, create a mongoDB Atlas database, get OAuth credentials, and visit the
github repository
Challenges we ran into
Figuring out how to store the images was a big challenge. This was the first time I really worked with mongoDB, so it took me a while to decide to store images in base 64. Additionally, setting up the routes for the database using Express was more complicated than I expected, so I had to spend a lot of time on that.
Accomplishments that we're proud of
I am very proud of putting together a full working website. Actually finishing the project in the allotted time was difficult, but it was well worth it.
What we learned
I learned how to make a proper full stack web app using Javascript.
What's next for crowdata
Crowdata can start integrating actual payments and percentages to make a revenue and become an actual business. Also, allowing other types of data to be requested on the platform like videos and audio is the next step. This way, crowdata can expand to more users and make machine and deep learning easier for all.
Built With
axios
bootstrap
css
express.js
google-gmail-oauth
javascript
mongodb
mongoose
node.js
oauth
react
Try it out
github.com | crowdata | Making data accessible to anybody for deep and machine learning! | ['Remington Kim', 'Valerie Kim'] | [] | ['axios', 'bootstrap', 'css', 'express.js', 'google-gmail-oauth', 'javascript', 'mongodb', 'mongoose', 'node.js', 'oauth', 'react'] | 100 |
9,899 | https://devpost.com/software/the-warrior-returns | Inspiration
When one mentions the entertainment industry, most people would think about films and music. Many people watch the Oscars, Grammys, Golden Globes, MTV Video Music Awards, BRIT Awards, etc.
Of course, there is a lot of glitz and glamour in the film and music industries. But would you be surprised to learn that these two are not the top-grossing sectors in entertainment?
As a matter of fact, these two put together do not even match half the revenue the video game industry is earning. According to the latest figures, the video game business is now larger than both the movie and music industries combined, making it a major industry in entertainment.
This year, the global games market is estimated to generate US$152.1 billion from 2.5 billion gamers around the world. By comparison, the global box office industry was worth US$41.7 billion while global music revenues reached US$19.1 billion in 2018.
Consider the top blockbuster movie to date, Avengers: Endgame. When it premiered on April 16, it raked in over US$858,373,000 during its opening weekend. It even surpassed last year's Avengers: Infinity War, which generated US$678,815,482 in gross revenue.
But while these films received so much attention and hype from the general public, they failed to outperform the highest-grossing entertainment launch in history, Grand Theft Auto V’s release back in 2013, which earned US$1 billion in just over three days.
What it does
It is a Story Based RPG game.
How I built it
Built on unity3d, all the UI designs and elements are built on photoshop.
Challenges I ran into
Developing elements for graphics.
Accomplishments that I'm proud of
Being able to build it within the deadline.
What I learned
I actually learned quite a lot about the ups/downs that the gaming industry faces, what kind of games actually take the world by storm.
What's next for The Warrior Returns
Complete the game look for producers to work with, and release it to production.
Built With
c#
photoshop
unity | The Warrior Returns | Reality is Brutal. Its time to face it. | ['Vasa karthik'] | [] | ['c#', 'photoshop', 'unity'] | 101 |
9,899 | https://devpost.com/software/days-since-last | Inspiration
Inspiration came from when I wanted to track the amount of time from an event in a readable format. It needed to be able to be reset and shareable.
What it does
You can create a new counter when you login with your Google account, the visibility can be changed to public or unlisted. All counters are shareable and reset on everyone's devices when the owner of the counter resets it. You can assign notifications to be sent to the desktop when the counter exceeds a time limit or gets reset.
How I built it
It was built with React for the front end, and MongoDB and Node.JS for the back end with Express, with the help of TypeScript.
Challenges I ran into
Since the counters needed to be able to be reset, and multiple people could be viewing the same counter, I needed a way to reset the counter for anyone who's looking at the counter. I used Socket.IO for this and it worked out just fine.
Accomplishments that I'm proud of
I'm proud of how the actual counter display turned out.
What I learned
I learned how to use the Notification API, something I've never used and it made a nice addition to the project.
What's next for Days Since Last
Improving the home page and the profile page if I get around to it.
Built With
css
html
mongodb
node.js
react
typescript
Try it out
github.com | Days Since Last | A reference and reminder tool for viewing the amount of time elapsed since an event. | ['Hunter Parcells'] | [] | ['css', 'html', 'mongodb', 'node.js', 'react', 'typescript'] | 102 |
9,899 | https://devpost.com/software/escape-from-sarscov-2-jl2s60 | Inspiration
We were inspired by the current pandemic and wanted to make a game that teaches people about the disease.
What it does
We created this game as a way to educate people about the Novel Coronavirus pandemic in a way that is engaging and fun. Players explore a 3D world where they learn about the Novel Coronavirus. Along the way, they collect information about virus transmission, symptoms, and prevention.
How we built it
We created the 3D models through Blender. Some of the 3D models were from an open source (free3.com). We also found textures from Poliigon. We rendered the models with shading and other tools built into Unity. The game was implemented through Unity. Some objects were coded with C#.
Challenges we ran into
Our scope for this project was slightly too large. We did not have time to implement the character getting and displaying the symptoms of Novel Coronavirus. We also did not have time to generate the NPCs to interact with.
Accomplishments that we're proud of
For the time allotted, we got a lot done. We were able to create a beautiful 3D world that we were able to navigate and explore. We were also able to display many fact and a QR code to a facts sheets in our world.
What we learned
This is the first time all 4 of us have used the Unity engine. Many of us had also never made 3D models nor programmed a game anymore, so this was a great learning experience in learning how to program a game, create 3D models, and learn quickly within a deadline.
What's next for Escape From SarsCoV-2
In the future, we plan to allow people to select from different illnesses. We also plan to add randomly generated people that you can interact with. We also want to allow the player to interact with items and do actions that can lower or raise their chance of getting the disease. Another feature we want to add is to allow players to collect information cards on diseases.
Built With
blender
c#
c++
unity
Try it out
github.com | Escape From SarsCoV-2 | A 3D game to teach about the Novel Coronavirus in the pandemic | ['Lilian Chan', 'Maggie Haddon', 'TopBloke_'] | [] | ['blender', 'c#', 'c++', 'unity'] | 103 |
9,899 | https://devpost.com/software/following-the-ultralineamentum | title
in game example
Inspiration
I got inspired to make this by looking at the Oregon Trail re-made into something else on REPL. I decided to try and take the basis of that game and try to create something that people would understand today.
What it does
It's a game where you follow a car throughout the U.S. across a line known as the ultralineamentum line, or the longest line in the U.S.
How I built it
I built it in Java on REPL using a sing class and 3 static methods.
Challenges I ran into
I ran into problems with the RNG system and balancing out variables so that it doesn't seem too crazy when playing.
Accomplishments that I'm proud of
I'm proud that I created something and enjoyed creating it. I learnt a few new things and how I can use certain skills to create something bigger.
What I learned
I learned that if you put your mind to it you can create anything you really want to.
What's next for Following the 'Ultralineamentum'
I decided to see how it would go and maybe a a few new things or just re-work it so it runs smoothly. And maybe add a gui later on in the future.
Built With
java
Try it out
usacargame.blazingrod29.repl.run | Following the 'Ultralineamentum' | A quick, interactive game that allows the player to ride through the U.S. by following the ultralineamentum line. | ['Kayetan Jarzabek'] | [] | ['java'] | 104 |
9,899 | https://devpost.com/software/duck-duck-goose-043nfd | Please enjoy this very small and cute Canadian goose
What is this?
This is a text-based adventure style story that focuses on Honker the goose (as seen in the thumbnail) and his quest in search of his best friend, Ducker the rubber goose. This is set at the University of Waterloo campus, with four (main) possible locations to explore.
How did I make this?
I used
Ink
, which is a narrative scripting language made just for storytelling.
What up next?
Some next steps would probably involve having some images (read: doodles) of Honker as he goes on his quest. Creating this story was actually really fun, so I plan on making more interactive stories like this in the future.
Hear me out on this
It would take you at most 5 minutes to explore every possible piece of information, depending on how fast you read. Take your time! Explore the forest, infuriate the residents, create chaos to your heart's content (or at least, to the game's limitations). Think this as Untitled Goose Game but Canadian, shorter, and more-text-y.
The domain created for this is:
myemotionalsupportrubberduckyis.online
Built With
css3
html5
ink
inkstudio
javascript
json
Try it out
github.com | Duck Duck Goose | Help Honker the goose look for Ducker, his emotional support rubber ducky. | ['Yuxi Qin'] | [] | ['css3', 'html5', 'ink', 'inkstudio', 'javascript', 'json'] | 105 |
9,899 | https://devpost.com/software/far-friends-36krwz | Landing Page
Landing Page Responsive
User Profiles
Login-Page-Reponsive
Register-Page-Responsive
The Inspiration
The world's a very isolated place now as of the outbreak of COVID-19. Inter-Friends aims to connect people across the world, because
we're all in this together
.
The Project
We've created an application where users from around the world can connect with each other and keep socializing throughout this rough time. On the front page you can see the most recent users online and new users who just registered. Once signed up you can view each other's profile to find the best Inter-Friend for this pandemic.
The Building
Backend hosting using Heroku
Backend using Node & Express (Josh)
Frontend using React + Global State (Ian)
The Difficulties
This project has led to some mind-numbingly frustrating experiences which at the end of the day was actually a small fix. In addition to that, some of our teammates dropped out leaving only the two of us. This unexpected dropout caused us to change languages/frameworks for the backend mid-project. No hard feelings.
The Accomplishments
This was a large, ambitious project for us to complete. Perhaps too large, but what we did get done is an achievement in and of itself!
The Lessons
Global State Management within React
Authentication using Node
Password Encryption using bcrypt
Never try to build a social media app with 2 people in 36hours
The Future
We have many plans for the future of this project, including adding the ability to actually send messages to other users. We have prepared database tables to create this functionality, but still, need to do more.
Built With
express.js
node.js
postgresql
react
sequelize
Try it out
vigilant-wiles-d4bac7.netlify.app | Inter-Friends | Make it easy to meet new friends abroad! | ['Ian M', 'Joshua O.'] | [] | ['express.js', 'node.js', 'postgresql', 'react', 'sequelize'] | 106 |
9,899 | https://devpost.com/software/htne-cleared-media | Positive user sentiment analysis for Twitter
Negative hashtag sentiment analysis for Twitter
Opening page for Twitter sentiment analysis. Sample user card
Text summarization of two news articles from CNN and Fox. Article summaries vary per link!
HTNE-cleared-media
Our HackTheNorthEast Project uses NLP and AI to analyze traditional and social media content to help us better understand the media we consume.
This project was inspired by not only by the advancements in the relatively new field of NLP/AI applications but also the abundance of events that have taken place in 2020. With so much happening now, it can be hard sometimes to sieve through all the information coming at you. We want to help ease that burden.
Alex has wanted to work on the backend system for this project utilizing new technologies such as creating instances using GCP App Engine and deploying the Flask server onto it. Understanding how they are created and how the instance deals with the files if say libraries lived only in a user's /usr/local/bin. Overall, diving into cloud technologies will only pave the road towards a larger skillset and a new way of thinking if say AWS or Azure came into mind.
Kynan wanted to work on improving his frontend skills in React and Material-UI. Understanding the patterns in creating a single page application will only improve him going forwards if he wanted to explore other frameworks such as Vue or Angular. The design and component aspects of React are growing and every project going forward will only be developed quicker and better.
Challenges
Backend
There were MANY challenges in the development of the backend. There was a classification algorithm using Naive Baye's Classifier that did not obtain results that we wanted. The data set was just not good enough to compare to the content mainly seen in Tweets (used reviews about Apple to build the model). You can see the remnants of the client used in the github repo.
Another challenge was the setup of the GCP App Engine instance and the libraries associated to NLTK. It took a lot of trial and error to understand how the
app.yaml
file and the instance worked, because the data collection within NLTK lived in the
/usr/local/bin
directory, and that did not necessarily existed in the instance. It was learned later on that it lived in the same directory as the app.yaml file and I had to imitate the folder structure of the NLTK collection for it to then be detected!
Being new to ML/AI/DS in general, I referenced great articles in the area of text summarization and sentiment analysis. An example is from
this blog
that followed the simple steps of NLP to help summarize groups of text.
Frontend
Design and playing around with the UI is always difficult without using CMS like WordPress. Conveying the information in a very clean way with great user experience is always difficult as we are not UX Designers by heart. Material UI is very aesthetic given that the colors and design worked out of the box by themselves. But customizing them to fit our needs was very difficult. There are always new ways of combining different css properties and nesting elements properly that makes the work easier and the final product better.
Cleared Media
Our client uses the following technologies:
React
Material-UI
The main goal of the project is to both showcase how people talk on social media but also how traditional media (e.x FOX and CNN) take about issues compared to each other. Providing some added clarity on how we consume media in our everyday lives
Twitter
Tweets are short posts and straight to the point. It is here where both popular and influential figures in our world can express their thoughts to their followers, but what kind of image are they trying to portray? Social media platforms, like Twitter, are playing a larger role in how we communicate and share important events.
Natural Language Processing is a growing field in AI where machines try to learn and understand human language and extract important details about it. In this example, we practice
Opinion Mining
, or
Entity-Sentiment analysis
. Using Google's Natural Language API to extract information about a user's tweets, we can analyze their stance online. Based on their scores, we can classify this user or tag to be generally positive, negative, neutral, or mixed online.
With this information, we would be able to attempt to extrapolate different aspects of the users given their information. For example, if one user was overly negative towards their current state of government, AI could determine their position in the political compass given enough training to classify what is left or right.
This program is setup in a way the following technologies are used in the server:
GCP App Engine: Host online to process tweets and the AI
GCP Natural Language API: Determine sentiment
Flask: Handling requests
NLTK: Collection of data to help pre-process tweets
The following is required to host your own server:
Twitter Developer Key
GCP Account and a GCP project
Service Account Key to allow administrative tasks to be done in the server
Optional: If you do not use GCP App Engine or GCP Compute Engine to run your server securely, you will require an API key to allow usage of GCP Natural Language API
As the data is sent back to the client, it will display the user's general sentiment score and details in regards to it.
DISCLAIMER: The analysis being performed DOES NOT confirm any bias or views towards political issues or current events and they DO NOT accurately represent any user/hashtag position or thoughts in their entirety.
CNN / FOX
It is not uncommon for traditional news sites like CNN and Fox News to cover the same event with different biases. It can be difficult at times to see the subtle differences between them. How they may frame certain events, or how they analyze them. This specific part of the app allows users to scrape CNN / Fox News articles and provide a side by side summary (created by an NLP algorithm) of how these news sites talk about issues for the user to read. Allowing the users to see different issues in different lights and decide where they stand on it.
Local Setup
A live server won't be active follow shortly after as this is hackathon project. To recreate this project ensure these steps are followed. As of now. This program is using Python 3.7
Create a virtual environment i.e.
python3 -m venv venv
in the directory with the client/server
Navigate to the server directory and install all dependencies required using
pip3 install -r requirements.txt
Create your own credentials to use with Twitter by following their developer portal. Note that you must create an .env file with the keys stored for security purposes.
Greate a GCP account/project and obtain an API to add into the project. If you use App Engine for server purposes, you must store your environment variables within
app.yaml
. Refer to GCP guides for the proper syntax
If you are only working towards locally hosting the server, create a service account and export GOOGLE_APPLICATION_CREDENTIALS to the service key. Follow the guidelines in GCP to get started.
Run the Flask server
python3 main.py
. The server should then be available on
http://localhost:8080
or
http://127.0.0.1:8080
To run the client:
Navigate to the
/client
folder
yarn
and then
yarn start
yarn build
for a production level build for your use.
NOTE
Many of the functions available are assumed to be correct input due to time limitations. As well, there are many lines of code that can be refactored and removed but due to the same reason above, will not be done.
Built With
app-engine
artificial-intelligence
beautiful-soup
flask
google-cloud
javascript
machine-learning
material-ui
nltk
python
react
tweepy
Try it out
github.com | ClearedMedia | In an era where information is so abundant, there are times when it feels overwhelming. We wanted to make a platform that can help us better understand the media we consume. | ['Alex Dong', 'Kynan Ly'] | [] | ['app-engine', 'artificial-intelligence', 'beautiful-soup', 'flask', 'google-cloud', 'javascript', 'machine-learning', 'material-ui', 'nltk', 'python', 'react', 'tweepy'] | 107 |
9,899 | https://devpost.com/software/money-converter-ia943q | My Money Converter
Since I am a seventh grader, and only have knowledge within Scratch's boundaries, I decided to solve a simple problem within my coding capacity.
My simple money converter, is capable of converting different types of currencies.
I used Scratch.
When I was coding the AI, I had many difficulties in getting the computer to solve the calculations correctly. I had to restart a few times, but pulled through and our money converter turned out great!
I learned how to code using AI.
My next steps for Money Converter are to add more currencies as this is just a start.
Built With
scratch
Try it out
scratch.mit.edu | Money Converter | A simple tool used to convert different currencies with ease. | ['Kishan Sharavanan', 'Maximilian Wiens Rintha'] | [] | ['scratch'] | 108 |
9,899 | https://devpost.com/software/finpro | Inspiration
It turns out that 32 percent of young adults have limited money management skills, according to a University of Illinois study.
Track your spending, individuals and companies can use this simple user friendly financial management app to track daily, monthly and yearly spending and expenses
What it does
Finpro is a financial management solution that helps individuals and businesses manage and track their expenses, invoices, budgets and card payments
How I built it
Backend will be built using AWS, Nodejs
Frontend is built using HTML/CSS, JavaScript, react, redux, bootstrap
Challenges I ran into
The back-end and limited time
Accomplishments that I'm proud of
I consider every aspect of the project as an accomplishment as it was new learning every step of the way.
What I learned
I became more familiar with languages and tools used to build it
What's next for FinPro
A lot.
• Good UI design
• Working features of scan, invoices and card payments
• Back-end implementation
• Training development to the development and management of the app
• Support for tools and resources
Built With
amazon-web-services
bootstrap
css
html5
javascript
react
redux
Try it out
finpro-mng.netlify.app | FinPro | An expense tracker web app | ['Queendalin Nduka'] | [] | ['amazon-web-services', 'bootstrap', 'css', 'html5', 'javascript', 'react', 'redux'] | 109 |
9,899 | https://devpost.com/software/verdict | Our logo
The first page where reviewers can input text data for analysis.
The output page where reviewers can see the most negative sentences along with keywords and a recommended decision.
Inspiration
It is an unfortunate truth that online gaming has a reputation for hostility. In a largely consequence-free environment inhabited by mainly anonymous men, it is not uncommon to see everything from racist comments to targeted harassment. For example, in League of Legends, the most played multiplayer PC game in the world, top players and streamers have recently spoken out on the declining state of the community and the concerning lack of punishment for toxic behavior, “There's no point in reporting people because Riot won't ban them, no one is scared of getting in trouble, no one is scared of getting banned.”
Unfortunately there is no good solution to this problem. Millions of people play online multiplayer games every day, and as a result video game developers can receive thousands of reports every day—each of which needs to be carefully assessed to give a fair punishment. However, there are still steps that can be taken to not only process reports quicker, but also maintain a fair punishment of reported players.
We created Verdict to achieve exactly this, as it uses Google Cloud’s natural language processing API to analyze player communication and streamline the process of assessing reports. In the end, a human will still review the report and make a final decision, but using sentiment and entity analysis, Verdict highlights both keywords and the most negative portions of a given chat log, as well as an overall score for the player’s communications and a recommended final decision. We hope that using Verdict, game companies can respond fairly and quickly to toxic behavior and foster a better multiplayer gaming experience for everyone.
What it does
Verdict offers the option to upload a chat log or the transcribed speech of a player onto its website in a text format. After that, it reports the overall sentiment of the text, a value between -1 and 1, and classifies the player’s communication as either toxic or not toxic. Furthermore, it also reports the five most negative sections of text where the lowest sentiment was found, providing reviewers a basis upon which they can assign a punishment. Lastly, it also highlights key words from the text, which could include profanity, racist statements, or other extracted entities.
How we built it
Backend:
Verdict uses Google Cloud's natural language processing API and sentiment analysis to assign a value to both each line of the text as well as the text as a whole. Using this data, we recommend a final decision as well as display the five most negative lines. Furthermore, we use the API to perform entity analysis on the text as a whole and extract meaningful phrases for reviewers to see. This data is then stored into a JSON and sent to the frontend for display.
Frontend:
The frontend of Verdict is built using React, which helps us create text boxes and buttons, as well as layout the final data. After the user inputs text data, it is sent as a POST request to a Python server hosted using Flask. The Python script then unpacks the data in the request and processes it in the backend to produce a JSON of the final data. This JSON is then sent back to React and it updates the frontend with the new data from the query.
Challenges we ran into
For many of us, this was our first experience in web development, and learning how to connect the frontend built using React and the backend built using Flask was difficult. Furthemore, this was our first time using natural language processing and sentiment analysis and accurately predicting toxicity from chat logs was difficult. Lastly, we had to learn many tools for frontend development, from using CSS to creating React apps.
Accomplishments that we're proud of
We’re proud of not only being able to make a reliable Python script for determining toxicity from chat logs, but deploying it on a server and making it eventually accessible online. We all enjoy playing video games and are proud of creating a tool to reduce toxicity and harassment in the gaming community.
What we learned
We learned how to make calls to the Google Natural Language Processing API and process the results in a meaningful way. Furthermore, we learned how to use Flask to deploy a Python server as well as use React to construct an intuitive and user-friendly UI.
What's next for Verdict
Unfortunately, online harassment and toxicity exists not just in the gaming community, but also in social media as well as in online forums and blogs. While the majority of content online is benevolent, we hope to expand Verdict to other online communities to both increase user retention as well as provide a safer and more inclusive experience.
Built With
google-cloud
javascript
natural-language-processing
node.js
python
react
Try it out
github.com | Verdict | A machine learning tool that analyzes player communications in order to allow a quick and fair response to toxic behavior. | ['Anthony Zhou', 'Cody Wang', 'Mathew Han', 'Auston Lin'] | [] | ['google-cloud', 'javascript', 'natural-language-processing', 'node.js', 'python', 'react'] | 110 |
9,899 | https://devpost.com/software/hack-the-northeast-2020 | Inspiration
I never made a game before and wanted to start off with something simple. I've played and enjoyed a few visual novels in the past and I thought they would be a good place for beginners like me to start off.
How it was built
I used the Ren'Py visual novel engine that uses Python for scripting. Used my own computer setup as the images for the majority of the visual novel for which I also used Photoshop. Finally, I also used Atom as my preferred text editor ;)
Challenges I ran into
I had absolutely no idea where to start since I've never made a game before nor was I really familiar with Ren'Py and Python. Therefore, a lot of research was needed to find out how to make my idea come true.
What I learned
I familiarized myself with Ren'Py and the basics of Python.
Built With
photoshop
python
ren'py
Try it out
github.com | Hack the Northeast 2020 Hackathon Simulator | A visual novel that is about the process of making a visual novel for the Hack The Northeast 2020 Hackathon. | ['Matthew Forster'] | [] | ['photoshop', 'python', "ren'py"] | 111 |
9,899 | https://devpost.com/software/strayalerthtne | Stray Alert
Authors: Bruce Choe, Gage Christensen, and Dylan Wong(currently not showing up as a contributor,
even though he made commits, see his github:
https://github.com/Wong-Innovations
).
The Problem
Currently, if you encounter a stray pet on the street, you have to get close enough to see its tag
in order to get in contact with the owner. In the case of dogs, this can sometimes be difficult, as
the dog may be aggressive or they may run away when approached. In other cases, a person may
retrieve the collar info but the owner may be unresponsive, and the person may not have much time
to stick around. Or, in the worst case, the animal in question may not even have a collar, so the
owner is impossible to find.
The Solution
We decided to solve this problem by creating an app that would allow stray finders to send out
alerts about spotted animals and allow pet owners to receive alerts that may be relevant to them.
This allows owners to have the chance to get notified about a lost pet even if the finder could not
get close enough to retrieve their info, and also enables higher response rates through dedicated
notifications. Key functionalities required for this app include:
Ability for a user to take a picture of a stray, and then send out an alert with a message and the
user's location. All user's within a valid range would receive the alert.
Ability for a user to register their pet with the app. Then, if an alert contained information
such as dog breed, the app would check that information with the user's profile to see if the alert
is more relevant to them. In addition, if the alert contained info such as phone numbers or
addresses from the collar, a high level notification that confirms that the owner's pet is found
would be sent. This increases the likelihood that a finder will be able to get in contact with the
owner.
Our Progress
We all came into this project with little Kotlin/Android experience. As such, we were not able to
produce a result due to various road blocks. However, we still think that this project is viable and
will continue working on it after the hackathon. The main functionality would be having a server
find the users relevant to an alert and then send the alert, and this will not be complicated to
implement. Funding could be acquired through a lightweight ad model that advertises pet products,
and any generated revenue could be donated to animal welfare causes. Currently, we have the
following:
UI prototypes.
Basic client-side functionality to determine whether an alert is relevant to the app owner.
Built With
kotlin
Try it out
github.com | StrayAlertHTNE | Android app for finding owners of stray pets, submitted to Hack the Northeast. | ['Gage Christensen', 'Bruce Choe'] | [] | ['kotlin'] | 112 |
9,899 | https://devpost.com/software/blackjack-rl | Inspiration
We wanted to implement Reinforcement Learning with web technologies.
What it does
You can play against a dealer that has been trained using Reinforcement Learning.
How we built it
We used
Parcel
to compile the website down to static files.
We used
TensorFlow.js
to train the computer.
Challenges we ran into
None so far
Accomplishments that we're proud of
We used TensorFlow.js.
What we learned
We learned TensorFlow.js.
What's next for Blackjack RL
Built With
html5
node.js
parcel
sass
tensorflow
typescript
Try it out
github.com
undergraduateartificialintelligenceclub.github.io | Blackjack RL | Training a computer to play blackjack. | ['Giancarlo Pernudi Segura'] | [] | ['html5', 'node.js', 'parcel', 'sass', 'tensorflow', 'typescript'] | 113 |
9,899 | https://devpost.com/software/kinetic-energy-generator | model on a full body
model on a leg
Sketch of what it would look like
Inspiration
I was inspired to think of the Kinetic Energy Charger because its a device that could potentially help people in more poverty stricken countries by helping them make electricity to power their small devices. Also, It creates renewable energy from motion, so people who exercise and care about the environment would like it.
What it does
It would generate power from movement
How I built it
I first sketched it out, then I used Tinkercad to arrange the parts
Challenges I ran into
I had trouble settling on how I wanted the device to generate energy. I settled on Magnetic induction.
Accomplishments that I'm proud of
the concept is solid, I know howto 3d model now. i was able to use echoAR to see how the model would look in real life.
What I learned
I learned a lot more about how generators work and generate electricity
What's next for Kinetic Energy Generator
3-d printing and perhaps assembling the generator
Built With
chrome
echoar
tinkercad
Try it out
www.tinkercad.com | Kinetic Energy Charger | Want energy on the go? Need a reliable source of energy derived from wasted motion?The K.E.C. (Kinetic Energy Charger) is just what you need! | ['Natural Taylor'] | [] | ['chrome', 'echoar', 'tinkercad'] | 114 |
9,899 | https://devpost.com/software/graphene-evfaqu | Graphene
The Problem
I am a high school student who is interested in computer science and competitive programming. In competitive programming, many problems are solved by representing the input data as a graph, a data structure consisting of a series node objects connected by edges. Typically, the input data will be formatted like this:
n m
a b w
a b w
...
Where "n" is the number of nodes, "m" is the number of edges, and each line "a b w" represents an edge from node A to node B with weight W. In competitive programming, it is important to test your solution for correctness against data other than the provided sample input. This often involved drawing out a graph by hand, typing the corresponding input format, and finally testing it against my program.
The Solution
For my HTNE entry, I developed Graphene, a web application which allows the user to create a graph visually and then instantly copy the input data representation. As this app was designed with the competitive programming scene in mind, the focus was on simplicity and efficiency. I built Graphene using the p5.js, a Javascript library which allowed me to quickly create the interactive graphical experience I needed. However, before HTNE I had never even written a line of Javascript, which represented a challenge early on in development. Along the way, I learned a ton about web development, as it was an area of computer science which I had not thoroughly explored before.
Expanding Ideas
Graphene is a simple application with a specific use case, but the core idea has much broader applications. The fields of computer science, data science, and machine learning are filled with concepts and structures that are best understood visually (vectors, artificial neural networks, word embeddings, etc). However, programs take in this data as text. The core ideas behind Graphene could be expanded into a larger website which allows developers to visualize various forms of data and export them in a "code-ready" format. I believe this would improve the accessibility of the aforementioned fields by allowing developers to "see what their program sees."
Built With
css
html
javascript
Try it out
github.com | Graphene | Streamlined conversion from graph visualization to input data. | ['Nolan Clement'] | [] | ['css', 'html', 'javascript'] | 115 |
9,899 | https://devpost.com/software/augnav | Inspiration
What it does
How I built it
Challenges I ran into
Accomplishments that I'm proud of
What I learned
What's next for
Built With
stuff | asdf | asdf | ['Deepak Ramalingam'] | [] | ['stuff'] | 116 |
9,899 | https://devpost.com/software/htne-submission | Inspiration
Fun little game to play in quarantine
What it does
Cars spawn on both sides of the road and the goal is to dodge them
How we built it
We used the Panda3D graphics module and Python to help build it
Accomplishments that we're proud of
We are proud to have a final acceptable project to submit
What we learned
-Python
-Panda3D
Try it out
github.com | Dodgepanda 3D game | A game where bystanders try dodging zooming cars | ['Rishi Sinha', 'Venkat Kunaparaju', 'AD C', 'Krish Naik', 'Aditya Chakka'] | [] | [] | 117 |
9,899 | https://devpost.com/software/alz-vision-639nrj | Home Page.
Upload Page to upload and describe a memory.
Redescribe page to redescribe an uploaded memory.
Detailed statistics page.
Analytics page displaying graphs.
Alz.vision
Inspiration
Nearly 50 million people worldwide fall victim to memory impairments such as dementia. We personally have met people who struggle with Alzheimer's and have forgotten critical information and cherished memories, even going as far as forgetting a loved one. And according to a report from Alzheimers Disease International, nearly 75% of Alzheimers patients worldwide go undiagnosed. Currently, when doctors diagnose dementia, they lack concrete data on a patient’s decline of memory, relying on a combination of brain scans, memory tests, and interviews with family members.
What it does
We decided to create Alz.vision, a web application that uses machine learning to help potential Alzheimers and Dementia patients. The app has 3 core components.
First, it prompts users to upload memories each day with a single sentence describing the memory.
Second, as time passes, the app prompts users to describe memories they have uploaded in the past again. Each of these descriptions are analyzed by a machine learning algorithm and are assigned scores.
Using state-of-the-art machine learning algorithms, such as Random Forest regression, Support Vector Regression, and clustering to name a few, the app analyses the scores and displays compelling graphs and statistics in real time for users and their doctors to view
By using machine learning to analyze many memories over time, Alz.vision is a data analysis tool that analyzes a user’s memory for signs of memory loss and provides doctors with key data to make a more accurate and informed diagnosis. The user uploads photos and videos to the application, along with a description of the event portrayed. As the user continues to upload photos and videos, they will be prompted to recall the memories by writing another description. Our similarity algorithm will analyze the two descriptions to determine their similarity. Using this data and algorithms such as Random forest regressions, Support vector regressions, Linear Regressions, and Natural Language Processing, the app provides graphs and visual aids to show a user's memory decline. In addition, it will search for any potential outliers in their memory loss and common keywords associated with those outliers.
Overall, by analyzing and performing data analytics on the user’s memory over time, Alz.vision is able to use state of the art machine learning algorithms to detect powerful trends as well as create compelling and easy to understand graphs, helping users take a more active role in their health and helping doctors make a better and more informed Alzheimer’s and dementia diagnosis. With more testing of our algorithm, we plan to expand our application to warn users if their declination of memory suggests Alzheimer’s or dementia and recommend that the user visits a doctor.
Accomplishments that we're proud of
We got the entire web application working together! We were really proud of how our application is currently functional and accurately creating graphs in response to user descriptions in real time. We can successfully upload memories and display them for redescribing. Also, our machine learning analysis is integrated into our app, so we're really happy about how everything is coming together. We should be ready to pitch our app and put it out to production.
How we built it
We used Flask, HTML, CSS, JS, and Bootstrap for the frontend and backend for this project. We used Flask-Mongodb and MongoDB Atlas as our database to store user information, images, descriptions, and scores. For the frontend, we used HTML/CSS/JS with Bootstrap. We implemented a text similarity algorithm based on Levenshtein Distance, Linear regressions, Random Forest regressions, Support Vector Regressions, and Natural Language Processing to analyze the descriptions and create compelling and meaningful graphs for both patients and doctors.
Challenges we ran into and What we learned
It was particularly difficult for us to upload images through MongoDB and we learned a lot about images with MongoDB and that there was a library which handled MongoDB with Flask (we were originally using only the original MongoDB library). It was also difficult for us to figure out how to return images from our machine learning algorithms to display on the website. We finally learned how to convert the graphs to base64 images, which we could then display on the website using HMTL.
For both the outlier detection and the sentence similarity scores, we tested a few models before reaching our final decision on the model which worked best. It was our first time using MongoDB Atlas and mongoDB, so it took some time to learn about the API. It was also a little difficult to work together online, but we used Discord as our platform and made sure to periodically check-in on each other.
From a non-technical standpoint, our team also learned a lot about Alzheimer's. We spent two weeks researching the disease to learn more about how it is diagnose and listening to real user stories.
Business Model
Market
According to WHO, There are nearly 10 million new cases of Dementia per year worldwide approximately, furthermore according to Alzheimers disease international 75% of people with dementia have not received a diagnosis. This makes our total market size over 40 million people.
Revenue Model
In terms of our revenue model, we will choose to make revenue in 2 key ways: Advertisements, and by offering a premium subscription offering advanced data analytic features and providing greater insight into a user’s change in memory.
Competitive Advantage
Finally, the competitive advantage. Current methods of diagnosing Alzheimers and Dementia include Interviewing family members, and conducting memory tests and brain scans, but there is no concrete data. Alz.vision on the other hand, analyzes images and memories over time to measure the change in memory, Uses ML and neural network to detect trends and patterns, and most important, provides concrete data.
Next Steps
We hope to share our app with local doctors to get their feedback on our app. We will adjust accordingly and then continue the design process to create a finished product. We especially want their feedback on how to display the data which will be most convenient to them. Then, we will pitch this product to local hospitals and clinics and hopefully collaborate with them to make this product a tool for patients to collect data to help doctors to diagnose dementia and Alzheimer's better. Overall we hope to see if we can make it a startup and push it out for our community and the world to use.
Built With
bootstrap
css3
flask
html5
javascript
machine-learning
mongodb
natural-language-processing
ntlk
numpy
python
sklearn | Alz.vision | A web application that analyzes the memory of users to determine signs of memory loss and provide key data for their doctors to make a more accurate and informed Alzheimers diagnosis. | ['Veer Gadodia', 'Shreya C'] | ['Finalist Prize'] | ['bootstrap', 'css3', 'flask', 'html5', 'javascript', 'machine-learning', 'mongodb', 'natural-language-processing', 'ntlk', 'numpy', 'python', 'sklearn'] | 118 |
9,899 | https://devpost.com/software/blackhole-kfwqr1 | App icon
Inspiration
We want to help people with mental stress at this difficult time, and provide a space for them to let go of their negative feelings.
What it does
Blackhole is a mental health app that provides the user for them to text their negative thoughts or any unwanted emotions that they are experiencing. Once they are done writing what they feel, they can press the ‘let it go’ button and the text is erased. This provides students with a safe outlet for them to channel their emotions -and be aware of what they are. It also physically allows students to let go of the negativity or bad thoughts they may carry. And, if there is any sensitive message, such as the user showing potential to suicide, it will pop a message suggesting to call emergency contact or professionalist for help.
Challenges we ran into
Time zone, my cat was meowing loud
Accomplishments that we're proud of
The function of the app is to help people, which we are really proud of. Also, the design looks mature.
What we learned
We focused on simple app functions to illustrate solutions for mental health problems. We learned how to connect the original phone function to the app and put animations in the app.
What's next for Blackhole
For the next step, we will focus on working on my accurate danger detection, and send the information of the user to a database or hotline for them to reach out and help the user. And add more features to the app to create a more fully implemented mental health app. We will try to create the university mental health app and implement the Blackhole there.
Built With
dart
firebase
flutter
Try it out
github.com | Blackhole | The mental health app provides the user for them to text their negative thoughts or any unwanted emotions that they are experiencing. | ['Nikita Morozov', 'ying Zhu'] | [] | ['dart', 'firebase', 'flutter'] | 119 |
9,899 | https://devpost.com/software/skillroad | Logo
Inspiration
Whether you want to run your own company or snag the corner office at your current job, you probably have an end-goal for your career. But do you really know what you need to do to get there?
You are unlikely to start charting a course, unless and until you know the destination. The first and foremost step in your career plan is to create a detailed vision statement comprising different positions or things that you wish to achieve at various points in the coming years.
What it does
SkillRoad
is a career advisor platform that allows our users to be ready for their next career level. It is your guide to finding the appropriate career path for you.
SkillRoad parses our users' skills. Then, it recommends different roadmaps for the next career level based on their skills. Finally, we provide our users with suitable courses to achieve their goals.
At SkillRoad we use Artificial Intelligence to analyze your actual skillset and find an enhanced career path appropriate for you! Feel free to take a look at our interactive Home page, where you can get a trial of our services.
How we built it
Our website has advanced machine learning and AI algorithms to make sure that you 'll identify the gap between your skills and market needs, and it works as follow:
When the user uploads his/her CV the first model parses their skills and previous experiences via NLP techniques.
The second model predicts the relevant skills and careers by using word2vec.
With this new skill set, we predict the next career level and show it to users.
we use python for creating the ML models and javascript for the website. Then we integrated all these things with flask and react. Also, we used bootstrap for the themes.
Challenges we ran into
The biggest challenge we faced was that the team worked together remotely, spread over four different time zones. Moreover, it was difficult to:
Create a complete Machine learning web application using React and Flask.
Developing the application workflow to be fully automated.
Dealing with different APIs and data types.
Accomplishments that we're proud of
This is our second time as a team and we have succeeded again to create a prototype during the Hackathon. Also, our model has been deployed as a real-life project.
What we learned
Deploy Flask with React.
Flask.
Teamwork
Deploy ML model on Google cloud
What's next for SkillRoad
SkillRoad will help anybody who is looking for a new job in their field, looking for improving their actual and future skills, and looking to get acquainted to some ideas that they may not have known before. Furthermore, it would display recommend courses, useful material, and open-source projects perfect for strengthening your skills.
To continue our project we will look for people in that field that if our users have any questions they can reach out to get any answers that they desire. We hope that SkillRoad will help to improve our skillset and know our right job path.
Built With
bootstrap
flask
github
python
react
Try it out
skillroad.tech
github.com
skillroad.uc.r.appspot.com | SkillRoad | Career advisor platform that allows our users to be ready for their next career level. | ['Mohamed Amr', 'Juan Alegría', 'Shayaan Akbar', 'G R'] | [] | ['bootstrap', 'flask', 'github', 'python', 'react'] | 120 |
9,899 | https://devpost.com/software/arcademy | Inspiration
Due to COVID-19, there has been a significant rise in remote education along with a rise with its challenges; it's hard for some children to be stimulated at home, especially being away from their classmates; this may affect their education, cause them to learn less, or have bad grades, which can add stress to parents. ARCademy tries to remove stress for both child and parent by offering interactive AR Companions to help your child be stimulated throughout the learning process, by motivating children through a point and reward system, and by customizing plans to your child's needs.
What it does
ARCademy offers a learning system based on what parents want to improve for their children. It keeps your child motivated to learn through a point system and AR companions and offers parents transparency on what the child has progressed through and what they still need to work on. It's a win-win scenario for everyone.
Business plan
Users will pay for the application; it's a subscription based plan. As parents are always looking for new ways to engage their child into learning, this app would be a definite buy from them. Our target audience is young with ages ranging from 5 to 12 years of age.
How we built it
The AR pets are built by echoAR and the app is made in Flutter. We use MongoDB for the backend to store user information.
Challenges we ran into
We had some trouble with the AR when it wasn't rendering on mobile
We lost a team-member
Adobe XD could not properly compile the UI into a Flutter mobile application
Accomplishments that we're proud of
Learning how to make echoAR mobile app with Unity and Flutter.
What we learned
Unity + AR (echoAR) + Flutter mobile application
What's next for ARCademy
We realize the challenge of everyone not having good internet, so making this application to be used for offline purposes is what we will do next.
Built With
adobe-xd
c#
dart
echoar
flutter
mongodb
unity
Try it out
github.com
github.com | ARCademy | Help your child lift-off in academics with ARCademy | ['Andrew Yang', 'Sohil Athare'] | [] | ['adobe-xd', 'c#', 'dart', 'echoar', 'flutter', 'mongodb', 'unity'] | 121 |
9,899 | https://devpost.com/software/smart-bot-75mfgo | Smartbot
Inspiration
Assist people conduct self checks and carryout buying and selling of goods and services easily.
What it does
It recognizes key words and give response
How I built it
Built with IBM watson assistance
Challenges I ran into
Implementing in a website or a social media app
Accomplishments that I'm proud of
Finally able to implement.
What I learned
How to work effectively with IBM watson
What's next for Smart bot
Making it accessible through USSD codes offline.
Built With
ibm
ibm-watson
Try it out
web-chat.global.assistant.watson.cloud.ibm.com | Smart bot | A chatbot system that assit individual conduct checks with just some key words, help them contact help centers and also assit them in marketing their goods and sevices | ['Ugochukwu Nnachor'] | [] | ['ibm', 'ibm-watson'] | 122 |
9,899 | https://devpost.com/software/labxr | Inspiration
Monolithic banks and financial institutions cater to corporations and wealthy businesses rather than typical people.
The goal of UNI is to provide financial empowerment and autonomy to individuals.
What it does
UNI is a personal decentralized financial banking system for the individual.
Instead of putting your money in a central institutionalized bank, your money exists as a secure versatile digital asset that only you can control.
A UNI account consists of
(1) a digital wallet that holds conventional currency [USD] and stable cryptocurrencies like Bitcoin
(2) a mobile / web app that produces a virtual card capable of debit card financial transactions
How I built it
Challenges I ran into
Accomplishments that I'm proud of
What I learned
What's next for UNI
The creation of a token / cryptocurrency to incentivize the adoption of UNI.
Built With
firebase
flutter
Try it out
bitbucket.org | UNI | UNI is a personal decentralized financial banking system for the individual. | ['Warp Smith'] | [] | ['firebase', 'flutter'] | 123 |
9,899 | https://devpost.com/software/the-long-night-a-blockchain-based-quiz-game-on-got | Inspiration
We happened to play few blockchain-based betting games recently and game development has always been in our list to study and we were working on that lately. When we got an opportunity like this, we decided to give our best try on this.
What it does
This is a Game of Thrones based multiplayer quiz game on the Ethereum platform. The players will be playing the quiz game until one player is out of lives.
How we built it
Challenges we ran into
Accomplishments that we're proud of
What we learned
What's next for The Long Night
Built With
ethereum
express.js
node.js
phaser.js
redis
socket.io
solidity
web3
Try it out
github.com | The Long Night | A blockchain based quiz game on Game Of Thrones on Ethereum Platform | ['Mekha Krishnan', 'Dennis Sam'] | [] | ['ethereum', 'express.js', 'node.js', 'phaser.js', 'redis', 'socket.io', 'solidity', 'web3'] | 124 |
9,899 | https://devpost.com/software/project-rcym3h | Inspiration
Over 30 million small businesses in the U.S. employ nearly half of the American private workforce. These small businesses range from catering and clothing shops to jewelry and restoration stores. But owning a small business isn’t always easy. Co-founder Samantha Su’s parents, for example, are in the jewelry retail business. They are unable to financially maintain advertisements in order to compete in Amazon’s big marketplace and have difficulty juggling texts of hundreds of customers over iMessage. In addition to daily competition with bigger companies to attract customers, small business owners have faced tremendous challenges in retaining and contacting customers during the recent COVID-19 pandemic.
What it does
Connectica is an online service app that connects local small businesses with consumers through one-on-one chats. Connectica easily bridges these businesses with their clients even when the clients may not visit the stores on a day-to-day basis. Not only does Connectica help business owners effectively build a loyal customer community, but also its chat function provides customers with personal experiences with the business owners, allowing a stronger and closer relationship to form between them.
How I built it
Connectica is built with Swift and Xcode. We used Cocoapods and based our project on existing Github projects with similar functionalities, then combined, refined, and integrated them into our own app.
Challenges I ran into
While developing, we had challenges with programmatically segueing between view controllers without the storyboard. Additionally, we have to set constraints without being able to visualize what's going on. Merging our chat function with the rest of our code was challenging because we had difficulties making both features of the code compatible with each other. Additionally, it was challenging to create an effective UI design that would be both simple to use and easy to navigate. We also struggled with narrowing the types of small businesses eligible to use our app because we want Connectica's users to access small businesses that may not have greater exposure to new customers otherwise.
Accomplishments that I'm proud of
We are proud of implementing a working chat function that allows businesses and customers to communicate with each other efficiently as well as connecting it to the rest of our backend development into a working app. We are also proud of our start-up design because we believe that it will make a great impact on millions of small businesses throughout the nation.
What I learned
We learned how to successfully combine the chat function code with other aspects of our app to create a useful app for our business. This hackathon also introduced some of our team members to Xcode and more of its features that can be used to create an iOS app.
What's next for Connectica
In the future, we will refine all the key features of Connectica and implement features such as an in-depth business profile page, reward system for users, and referral programs among customers. After making these adjustments, we will publish the first version of Connectica on the App Store in August. We will also make Connectica Android compatible by the end of this year.
Built With
firebase
swift
Try it out
github.com | Connectica | Connectica is an online service app that connects local small businesses with consumers through one-on-one chats. | ['Samantha (Ziyao) Su', 'Yi Xie', 'Vicky Liu', 'Vivian Chen'] | [] | ['firebase', 'swift'] | 125 |
9,900 | https://devpost.com/software/autonomous-drone-delivery-system-wdfvy0 | Drone's POV with the model
Check graph title(shows accuracy of our model)
Math formulas we created to implement the probability model
An example of the object detection model MobileNet
Example of our data preprocessing
The architecture for our steering deep learning CNN model
The progression of our MSE over 25 epochs of training
See caption
Inspiration
The COVID 19 pandemic has affected all of us. Living in one of the epicenters of the outbreak in the US has been eye-opening. When we take a walk outside, no one out on the streets, playing in parks, or living their lives. Life at home has been even harder: buying groceries is frightening, purchasing necessities from online sites takes ages, and the entire online delivery system is strained and on the verge of collapse. People who are sick and in self-quarantine can not even exit their house due to the highly contagious nature of COVID-19. Likewise, the healthy and elderly are too scared to exit and buy essentials, such as food, due to fears of getting sick. Our project’s goal is to incentive everyone to stay home by making life a little easier and convenient. We want to return to normalcy as quickly as possible, and our method to aid social distancing, we thought, was the best way to accomplish this. Current autonomous drone navigation systems are neither robust or adaptable due to their heavy dependence on external sensors, such as depth or infrared, and we decided to take a deep learning approach and tackle both this and the COVID-19 problem together.
What it does
Using deep learning and computer vision, we were able to train a drone to navigate by itself in crowded city streets. Our model has extremely high accuracy and can safely detect and allow the drone to navigate around any obstacles in the drone’s surroundings. We also developed an app that allows users to communicate with the drone and send his/her coordinates to a real-time database.
How I built it
To implement autonomous flight and allow drones to deliver packages to people swiftly, we took a machine learning approach and created a set of novel math formulas and deep learning models that focused on imitating two key aspects of driving: speed and steering. For our steering model, we first used gaussian blurring, filtering, and kernel-based edge detection techniques to preprocess the images we obtain from the drone's built-in camera. We then coded a CNN-LSTM model to predict both the steering angle of the drone. The model uses a convolutional neural network as a dimensionality reduction algorithm to output a feature vector representative of the camera image, which is then fed into a long short-term memory model. The LSTM model learns time-sensitive data (i.e. video feed) to account for spatial and temporal changes, such as that of cars and walking pedestrians. Due to the nature of predicted angles (i.e. wraparound), our LSTM outputs sine and cosine values, which we use to derive our angle to steer. As for the speed model, since we cannot perform depth perception to find the exact distances obstacles are from our drone with only one camera, we used an object detection algorithm to draw bounding boxes around all possible obstacles in an image. Then, using our novel math formulas, we define a two-dimensional probability map to map each pixel from a bounding box to a probability of collision and use Fubini's theorem to integrate and sum over the boxes. The final output is the probability of collision, which we can robustly predict in a completely unsupervised fashion.
Challenges I ran into
We faced many challenges while creating our project, and two of the main problems were controlling our drone with a computer and optimizing runtime of our models. As for interfacing, the Parrot Bebop 2 drone is meant to be controlled with a mobile app that the Parrot company officially designed. However, there was no clear-cut way to control a drone with our computer, and finding a way to perform this took a lot much more effort and time than we had ever anticipated. We attempted using many libraries and APIs, the main ones include Bebop Autonomy, Olympe, and PyParrot. We first tried using Bebop Autonomy, a ROS driver specifically for Parrot drones and found that it wasn’t compatible with either of our computers. Then we discovered Olympe, which was a Python controller programming interface for Parrot drones and computers. We were able to get this working, but we found that the runtime of its scripts took too long and the level of complexity of its scripts was a little too much for us to handle. Finally, we found PyParrot and compared to Olympe the scripts were much easier to write and the API was overall a lot more user-friendly with a wide range of examples. Not only that, but it was open source meaning we could directly edit built-in functions, which was extremely convenient.
Another major challenge we face is runtime. After compiling and running all our models and scripts, we had a runtime of roughly 120 seconds. Obviously, a runtime this long would not allow our program to be applicable in real life. Before we used the MobileNet CNN in our speed model, we started off with another object detection CNN called YOLOv3. We sourced most of the runtime to YOLOv3’s image labeling method, which sacrificed runtime in order to increase the accuracy of predicting and labeling exactly what an object was. However, this level of accuracy was not needed for our project, for example crashing into a tree or a car results in the same thing: failure. YOLOv3 also required a non-maximal suppression algorithm which ran in O(n3). After switching to MobileNet and performing many math optimizations in our speed and steering models, we were able to get the runtime down to 0.29 seconds as a lower bound and 1.03 as an upper bound. The average runtime was 0.66 seconds and the standard deviation was 0.18 based on 150 trials. This meant that we increased our efficiency by more than 160 times.
Accomplishments that I'm proud of
We are proud of creating a working, intelligent system to solve a huge problem the world is facing. Although the system definitely has its limitations, it has proven to be adaptable and relatively robust, which is a huge accomplishment given the limitations of our dataset and computational capabilities. We are also proud of our probability of collision model because we were able to create a relatively robust, adaptable model with no training data.
What I learned
Doing this project was one of the most fun and knowledgeable experiences that we have ever done. Before starting, we did not have much experience with connecting hardware to software. We never imagined it to be that hard just to upload our program onto a drone, but despite all the failed attempts and challenges we faced, we were able to successfully do it. We learned and grasped the basics of integrating software with hardware, and also the difficulty behind it. In addition, through this project, we also gained a lot more experience working with CNN’s. We learnt how different preprocessing, normalization, and post processing methods affect the robustness and complexity of our model. We also learnt to care about time complexity, as it made a huge difference in our project. Finally, we learned to persevere through the challenges we faced.
What's next for Autonomous Drone Delivery System
Despite the many challenges we faced, we were able to create a finished working product. However, we believe that there are still a few things that we can do to further improve upon. To further decrease runtime, we believe using GPU acceleration or a better computer will allow the program to run even faster. This then would allow the drone to fly faster, increasing its usefulness. In addition, training the model on a larger and more varied dataset would improve the drone’s flying and adaptability, making it applicable in more situations. With our current program, if you want the drone to work in another environment all you need to do is just find a dataset for that environment. Currently, the drone uses its own wifi signal to communicate with the computer. This means that the drone could only fly as far as its wifi range allows it to. If we could move all the processing from our computer onto the drone however, it would give the drone an unlimited range to fly in making it even more adaptable in situations. Our GUI for our mobile app is pretty plain, and we hope to later implement and create a better app with a better and easier to use design. Lastly, integration with a GPS would allow long-distance flights as well as urban city traveling.
A self-flying drone is applicable in nearly an unlimited amount of applications. We propose to use our drones, in addition to autonomous delivery systems, for conservation, data gathering, natural disaster relief, and emergency medical assistance. For conservation, our drone could be implemented to gather data on animals by tracking them in their habitat without human interference. As for natural disaster relief, drones could scout and take risks that volunteers are unable to, due to debris and unstable infrastructure. We hope that our drone navigation program will be useful for many future applications.
Built With
keras
numpy
pandas
parrotdrone
pyparrot
python
tensorflow
Try it out
github.com | Using Autonomous Drones to Deliver Supplies During COVID-19 | A novel deep learning and computer created based drone navigation system by Allen Ye and Anant Bhatia to aid in autonomous delivery during the COVID-19 outbreak. | ['Allen Ye'] | ['1st Place', 'Best Drone Hack', 'Featured on Website'] | ['keras', 'numpy', 'pandas', 'parrotdrone', 'pyparrot', 'python', 'tensorflow'] | 0 |
9,900 | https://devpost.com/software/covaid-53hv21 | CovAid Register Page
CovAid Login Page
CovAid Requests Page
CovAid Requests Viewer
CovAid Request Submission
CovAid Home Page
Inspiration
The world we live in has changed dramatically amidst the COVID-19 outbreak. Although some of us are safe at home with the proper equipment, a large portion of the population does not have access to essentials. In analyzing the issue, we realized the immunocompromised currently had no access to essentials as they could not simply leave their houses to go to a grocery store. We decided to provide a solution to this problem by creating a website in which we could allow users to make virtual requests for items, such as toilet paper or hand sanitizer, and then enable volunteers to accept these requests to donate supplies to them. As there is no preexisting platform that allows for direct pairings between users and volunteer deliverers, we believe this is the perfect solution to help those most impacted by COVID-19.
What it does
CovAid is a web application that connects volunteers to those in need during the COVID-19 outbreak using AI-driven intelligence. The website connects at-risk users with volunteers willing to donate necessities. Users can make requests for items to the website and volunteers can respond to those requests. These pairings are created efficiently with a machine learning algorithm that takes into account various factors such as the distance between the user and the volunteer.
How we built it
Through the development of CovAid, we were able to learn how to integrate Flask, JavaScript, and jQuery as our back-end with HTML and Bootstrap together to develop a website from scratch. We used SQL to operate the database of users and the Google API to calculate the miles and estimated time between users. These topics were new to us and we were able to truly learn how to integrate every part together to create a fully-functioning website. In order to perform the matching between users and volunteers, we developed a Machine Learning Neural Network model to sort the requests on a volunteer’s page, as we wanted requests most relevant to the volunteer to show up when a volunteer is searching for a request to accept. We used Keras, NumPy, Pandas, and a Sequential Machine Learning Neural Network model with Dense layers to develop our model before implementing it into our website.
Challenges we ran into
We faced numerous challenges when it came to properly communicating with Flask view and the various HTML templates. Since CovAid is a dynamic site form data had to be sent back and forth between the files and stored in a database. Using a database was something new to all of us and understanding how to integrate it for our needs was a major roadblock for a while. Another major challenge was implementing our machine learning sorting algorithm with our Flask and HTML to sort the requests for each volunteer, since we had to learn how to get live user data to enter into the model.
Accomplishments that we're proud of
We are proud of how we could efficiently push out a website while allowing everyone on our team to contribute equally. After beginning with our entire team working together to create the basic layout of our website, we split up into two teams. Shrey and Atin worked on the front-end and back-end of the website while Anirudh and Aarav worked on the machine learning aspect of the project. We also learned various CS skills while also helping our community at the same time. In addition, we are also pleased that we have created another scenario that AI can help ease our lives. We are excited to see how our project will be able to create opportunities for other people to make a positive impact on their surroundings.
What we learned
In developing CovAid, aside from exploring new software such as Bootstrap and Flask, we fully understood the broader impacts of our project — that any simple act of kindness can be influential, especially to those that are impacted the most from issues like these.
What's next for CovAid
In order to create a real difference in our community we hope for CovAid to be more widespread and have a larger impact on the world. We also want to implement a system in which users are able to be further interconnected. Our vision is that through our product everyone will have access to essentials and will stay safe as our world continues to change from COVID-19.
Built With
bootstrap
css3
flask
google
html
javascript
jquery
keras
machine-learning
numpy
pandas
python
sqlalchemy
Try it out
github.com | CovAid | CovAid is a web application that facilitates deliveries to those in need during these pressing times. The website connects at-risk users with volunteers willing to donate necessities. | ['Atin Pothiraj', 'Aarav Khanna', 'Shrey Gupta', 'Anirudh Bansal'] | ['2nd Place'] | ['bootstrap', 'css3', 'flask', 'google', 'html', 'javascript', 'jquery', 'keras', 'machine-learning', 'numpy', 'pandas', 'python', 'sqlalchemy'] | 1 |
9,900 | https://devpost.com/software/covid-19-estimation-online-test-based-on-lung-sound | Inspiration
Inspired from
https://voca.ai/corona-virus/
What it does
COVID-19 Sound Classifier
How I built it
Using ml5.js and p5.js function for classifying sound. Train data using the Teachable Machine
Challenges I ran into
Low accurate and has not been scientifically tested
Accomplishments that I'm proud of
This is an alternative approach to estimating COVID-19
What I learned
Sound classification can be used in other symptoms
What's next for COVID-19 Estimation Online Test Based on Lung Sound
Scientifically testing and built it in an embedded system
Built With
ml5js
p5js
Try it out
bit.ly
github.com | Covid-19 Estimation Online Test Based on Lung Sound | Covid-19 sound classifier project | ['Rachmad Imam Tarecha'] | [] | ['ml5js', 'p5js'] | 2 |
9,900 | https://devpost.com/software/pick-me-up-with-twilio | Inspiration
COVID-19 Anxiety since the epidemic is not only a issue of physical health, but an issue of mental health as well.
What it does
Pick-me-Up targets anxiety in young adults by providing a method of tracking and reflecting on the user’s experiences. We offer an easy way to document the time, place, and feeling of an anxious episode so the user can have a productive conversation with themselves and a therapist. Our product uncovers trends in a person’s anxiety through the habitual usage of our product. It is also easily tailored to the patient per the user or user therapist jurisdiction. This is done to ensure each user maximizes the potential of the program.
How I built it
We deployed web app on google cloud engine. The phone interaction was made possible by the Twilio API. (Huge Thanks to Lizzie)
Challenges I ran into
Twilio API
Secure Data Storage
Accomplishments that I'm proud of
Fast turn around
What I learned
Twilio module integration.
What's next for Pick-Me-Up
Make a real product.
Built With
express.js
google-cloud
javascript
mongodb
react
Try it out
github.com | Pick Me Up with Twilio | Pick-Me-Up offers an easy way to document the time, place, and feeling of an anxious episode so the user can have a productive conversation with themselves and a therapist about their mental health. | ['Ian Reyes', 'kurien12thomas', 'Emily Smith'] | [] | ['express.js', 'google-cloud', 'javascript', 'mongodb', 'react'] | 3 |
9,900 | https://devpost.com/software/internationale-forschungsplattform | Logo Forschungplattform
Diese Plattform ensteht in Kooperation mit der Crisis_Magament_Plattform
sodass dem Wissenschaftler, Forscher und Erfinder dieser Forschungsplattform
immer die aktuellsten Daten zur Verfügung stehen.
Beschreibung des Projektes:
Es handelt sich bei diesem Projekt um eine neue Art der Forschung und einem noch nie dagewesenen weltweiten Projekt. Diese Plattform verbindet Wissenschaftler, Forscher und Erfinder in einer Form, die es weltweit noch nicht gegeben hat.
Die Daten, die entscheidend für Forschung sind, werden International zusammengestellt und stehen allen Forschern, Wissenschaftlern und Erfindern dieser Plattform zur Verfügung.
Dies ist aber nicht nur ein Internationales wissenschaftliches Projekt, sondern beinhaltet gleichermaßen ein soziologisches Projekt. Es ist ein dreifaches wissenschaftliches Projekt, welches eine große Zukunft hat. Stellen Sie sich selbst die Frage, was Sie für Fortschritte in allen Bereichen der Forschung machen können, wenn mit einmal statt 200 Wissenschaftlern 50000 Wissenschaftler, Forscher und Erfinder zusammen an den Projekten forschen. Mit einem unglaublichen Wissensspektrum und Ideen gebündelt auf einer Plattform zum Wohle aller Menschen.
Die Planung und organisatorischen Maßnahmen, ergeben sich aus dem Verbund internationaler Wissenschaftler, Forscher und Erfindern. Diese organisieren mithilfe dieser Plattform virtuelle Meetings um gemeinsam zu Forschen.
Daten werden durch ein Expertenteam analysiert und in der Plattform eingebunden, sodass jeder auf dieser Plattform immer und jederzeit auf den neusten Stand der aktuellen Forschung ist. Die Nutzer dieser Plattform haben zudem die Möglichkeit eine bestimmte Seite zum Beispiel eines Forschungsinstitutes unserem Team vorzuschlagen und nach einer Prüfung durch ein Team und der Absprache mit dem Betreiber der Seite wird diese integriert,
zudem können Forschungseinrichtungen ihre Seite mit einbinden lassen, sodass alle auf der Plattform befindlichen Nutzer die Informationen der Seite sehen können.
Das Gute an dieser Plattform ist, dass die Nutzer sich selbst organisieren, Meetings planen können und sich jederzeit austauschen können.
Gleichzeitig haben Nutzer die Möglichkeit Ihre Entdeckungen international mit allen Nutzern der Plattform zu veröffentlichen.
Wer steht eigentlich hinter dieser ganzen Aktion?
Mein Name ist Michael Rhein, ich leitete die Patentverwertung TIZ-NORD Wilhelmshaven Technisches Innovations Zentrum für Forschung und Patentverwertung. Als die Pandemie losging, überlegte ich welche Möglichkeiten es gibt, um schnellstmöglich diese Krise zu bewältigen. So suchte ich am Anfang alleine nach Lösungen bis ich bemerkte, dass genau hier die Lösung zu finden ist. Aufgrund dessen überlegte ich wie ich es schaffen könnte, so viele Menschen wie möglich zusammenzubringen, um eine Lösung gemeinsam zu finden. Hierbei war mir aber von Anfang an klar, dass dies nur mit Experten auf dem Gebieten der Wissenschaften, der Forschung und mit den daraus resultierenden Erfindungen oder medizinischen Mitteln der Erfinder möglich sein würde und so entwickelte ich das Projekt der Forschungsplattform. Mir war bewusst, dass dies mich vor enorme Herausforderungen stellen würde und wird, doch diese werde ich gemeinsam mit allen zusammen bewältigen können.
Wie ist es mit den Rechten?
Ich habe lange überlegt wie man den rechtlichen Teil dieser Plattform Organisieren könnte. Jetzt habe ich auch hierzu eine Lösung gefunden. Alle die sich auf dieser Plattform registrieren, willigen einfach ein, dass die Rechte an den auf der Forschungsplattform gemachten Forschungsergebnisse, Erfindungen zum Wohle der Menschheit, auf alle auf der Plattform registrierten Nutzer die an der Erstellung dieser Erfindung, Lösung oder den Forschungsergebnissen gleichmäßig verteilt werden. Dies ist eine optimale Lösung um gemeinsam zu Forschen. Zur Umsetzung dieser Regelung Akzeptieren die Nutzer die AGBs der Plattform, in denen alle rechtlichen Punkte festgelegt sind.
Wie sieht es Finanziell aus? Wer Finanziert das ganze?
Es geht bei dieser Plattform nicht darum, den größtmöglichen Gewinn zu erzielen, vielmehr geht es darum gemeinsam Dinge zu entwickeln die eine solche Situation wie Coovid-19 vermeiden können oder zumindest eine schnelle Lösung dafür herbeiführen können. Ich versuche über politische Wege von den Regierungen aus den die Wissenschaftler, Erfinder und Forscher stammen finanzielle Fördermittel zu erhalten, diese werden dann an die jeweiligen Forschungsinstitute, Wissenschaftler, Forscher und Erfinder aufgeteilt, hierzu werde ich ein Finanz Team zusammenstellen.
Wir sind alle Menschen, wir Leben alle auf diesen Planeten, lassen Sie uns gemeinsam Technologien und Möglichkeiten entwickeln, die wir uns heute eventuell noch nicht einmal vorstellen können.
Diese Plattform lebt wie wir alle, von stetiger Veränderung und Anpassung. Gemeinsam entwickelt sich diese Plattform durch uns alle. Und so werden wir Teil dieser Plattform.
Description
This project is a new type of research and an unprecedented global project that connects scientists, researchers and inventors in a way that has never existed before.
The data, which is crucial for research, is compiled internationally and is available to all researchers, scientists and inventors of this platform.
However, this is not just an international scientific project, it also includes a sociological project. It is a triple scientific project that has a great future. Ask yourself the question of what progress we can make in all areas of research if, instead of 200 scientists, 50,000 scientists, researchers and inventors research the projects together. With an incredible range of knowledge and ideas Thum bundled on a platform for the benefit of all people. The planning and organizational measures result from the association of international scientists, researchers and inventors. With the help of this platform, they organize virtual meetings to conduct research. Data is analyzed by a team of experts and integrated into the platform, so that everyone on this platform is always up to date with the latest research. Users of this platform also have the opportunity To propose a certain page, for example a research institute, to our team and after an examination by a team and consultation with the operator of the page, this is intrigued, and research institutions can have your page integrated so that all users on the platform see the information on the page The good thing about this platform is that users can organize themselves, plan meetings and exchange information at any time. At the same time, users have the opportunity to publish their discoveries internationally with all users of the platform.
Who is actually behind this whole campaign?
My name is Michael Rhein, I managed the patent exploitation TIZ-NORD Wilhelmshaven Technical Innovation Center for Research and Patent Exploitation. When the pandemic started, I thought about the options for dealing with this crisis as quickly as possible. So in the beginning I looked for solutions on my own until I noticed that the solution can be found right here. Because of this, I thought about how I could bring as many people as possible together to find a solution together. However, it was clear to me from the beginning that this would only be possible with experts in the fields of science, research and the inventions or medical means of the inventors resulting therefrom and so I developed the project of the research platform. I was aware that this would and will present me with enormous challenges, but I will be able to master these together with everyone.
What about the rights?
I have long considered how to organize the legal part of this platform. Now I have found a solution for this too. All those who register on this platform simply agree that the rights to the research results made on the research platform, inventions for the benefit of mankind, to all users registered on the platform who are involved in the creation of this invention, solution or the research results are evenly distributed . This is an optimal solution for doing research together. To implement this regulation, users accept the general terms and conditions of the platform, in which all legal points are defined.
What does it look like financially? Who finances the whole thing?
This platform is not about making the greatest possible profit, rather it is about developing things together that can avoid such a situation as Coovid-19 or at least bring about a quick solution. I try to get political funding from the governments from which the scientists, inventors and researchers come, which are then distributed to the respective research institutes, scientists, researchers and inventors, for this I will put together a finance team. We are all human beings, we all live on this planet, let's develop technologies and opportunities together that we may not even be able to imagine today. This platform, like all of us, lives from constant change and adaptation. Together we all develop this platform, so we become part of it.
Built With
css3
deutsch
englisch
google-analytics
google-translate
japanisch
javascript
php
php5
russisch
spanisch
strato
weltweit-alle-spachen
zoom
Try it out
forschungsplattform.com | Internationale Forschungsplattform Research platform | für weltweit alle Forscher, Wissenschafer und Erfinder for everyone Inventor researcher and scientist | ['Michael Rhein', 'Carola Günther'] | [] | ['css3', 'deutsch', 'englisch', 'google-analytics', 'google-translate', 'japanisch', 'javascript', 'php', 'php5', 'russisch', 'spanisch', 'strato', 'weltweit-alle-spachen', 'zoom'] | 4 |
9,900 | https://devpost.com/software/national-vision-administrators-l-l-c |
window.fbAsyncInit = function() {
FB.init({
appId : 115745995110194,
xfbml : true,
version : 'v3.3'
});
// Get Embedded Video Player API Instance
FB.Event.subscribe('xfbml.ready', function(msg) {
if (msg.type === 'video') {
// force a resize of the carousel
setTimeout(
function() {
$('[data-slick]').slick("setPosition")
}, 2500
)
}
});
};
(function (d, s, id) {
var js, fjs = d.getElementsByTagName(s)[0];
if (d.getElementById(id)) return;
js = d.createElement(s);
js.id = id;
js.src = "https://connect.facebook.net/en_US/sdk.js";
fjs.parentNode.insertBefore(js, fjs);
}(document, 'script', 'facebook-jssdk'));
Website :
https://apps.apple.com/us/app/nva-vision-benefits-member-app/id1381492609
Address : 1200 Route 46 West, Clifton, NJ 07013
Phone : +1 973-574-2400
National Vision Administrators provides vision benefit programs designed to meet the needs of our clients. We have been providing flexible vision benefit options since 1979 and our provider network includes tens of thousands of licensed eye care professionals and optical retailers in all 50 states and Puerto Rico. Our President, David S. Karlin, RPh, believes that practicing mindfulness improves organizational wellbeing, reduces stress, and increases productivity. National Vision Administrators unique approach to management provides excellent customer service and support across our network. We are proud of our 99% customer retention history and believe it is the greatest testimony to our products and services. Check out our website to learn more how National Vision Administrators can improve the quality of your benefits package today! | National Vision Administrators, L.L.C. | insurance | ['National Vision Administrators, L.L.C.'] | [] | [] | 5 |
9,900 | https://devpost.com/software/covid-19-patient-non-patient-risk-prediction-using-ai |
window.fbAsyncInit = function() {
FB.init({
appId : 115745995110194,
xfbml : true,
version : 'v3.3'
});
// Get Embedded Video Player API Instance
FB.Event.subscribe('xfbml.ready', function(msg) {
if (msg.type === 'video') {
// force a resize of the carousel
setTimeout(
function() {
$('[data-slick]').slick("setPosition")
}, 2500
)
}
});
};
(function (d, s, id) {
var js, fjs = d.getElementsByTagName(s)[0];
if (d.getElementById(id)) return;
js = d.createElement(s);
js.id = id;
js.src = "https://connect.facebook.net/en_US/sdk.js";
fjs.parentNode.insertBefore(js, fjs);
}(document, 'script', 'facebook-jssdk'));
RiskPredictionTool GUI
Confusion Matrix
Receiver Operating Characteristics
Optimization for minmize the (1-Accuracy)
Predict the Result of Patient Risk Level
Data Flow Diagram
Inspiration
Today , important of human health , we need solve two major problems. First one spreading Risk Factors of Covid-19 we must need to find and reduce and Second one Risk Factors of Covid-19 patients. Our inspiration simply we need to overcome pandemic era.
Problem Identification
Last one month i had search valid data-set for Covid-19 patients. But the meanwhile i have understood most of the data-base used for tracing and counting. so We have to planned a common Machine Learning tool for Reduce the Corona Spreading level based on their foods, medical history, social distance parameters.
The Risk Prediction patients based on their clinical check list. we have to plan a common tool for Prediction.
What it does
It's a common Tool GUI, The name itself , if you have local area patient checklist/attributes then easy to Train the Data using this GUI and finally predict the Risk level of Unknown Patients.
Train Data(Risk Level Known Patient) --> Feature Normalization --> optimized Feature Selection ---> Train The Classifier
Test Data(Risk Level Unknown)--> Normalization---->selected Features---> Predict the Result with the help of Trained Classifier.
In simple-ways, if you have your local area patient check list from that we have find the Risk prediction/Infection Chance. Most importantly , if we do in correct way, we will easy to find which are the parameters like foods and social distance type etc reduce the Covid-19 pandemic.
How I built it
using MATLAB GUI . Design and Evaluate the Performance
Challenges I ran into
Original Dataset Collection is complex. In this Data-set we have achieved 87% Training Accuracy and More than 75 % Testing Accuracy.
Accomplishments that I'm proud of
Initial Version 1.0 Task Completed
What I learned
We have already entered Covid-19 pandemic , The Process like Reverse Engineering by using Machine Learning reduce the impact of Covid-19.
What's next for Covid-19 Patient/Non-Patient Risk Prediction using AI
here up-to now , we have completed Simple level ML Tool for Risk-class Prediction. In Future we need to update both Risk Class and Values by using Fuzzy and so many AI Techniques.
Built With
excel
gui
matlab
tool
Try it out
github.com | Covid-19 Patient/Non-Patient Risk Prediction using AI | For Patient/Non-Patient Covid-19 Risk prediction like Infection Chance or Health of Patients | ['AmburoseSekar SiluvaiRaj'] | [] | ['excel', 'gui', 'matlab', 'tool'] | 6 |
9,900 | https://devpost.com/software/virality-your-social-accountability-app | app icon
Inspiration
# Problem Statement
In a short duration of 4 months, the number of Corona-virus cases reached 1,925,877 and
still increasing day by day. It is a pandemic of respiratory disease spreading from person-
to-person caused by a novel corona-virus.
“As of 6 February 2017, a total of to 19217 hospitals and health-care facilities in 177 countries.”
“Available statistics show that over 30% of WHO Member States report to have less than 10 medical doctors per 10000 population.”
“Available statistics show that over 45% of WHO member states report having less than 1 physician per 1000 population.”
Even the world's largest democracy India only has 7,39,024 beds in 37,725 facilities and has 1 doctor for every 1457 citizens.
This stats clearly shows that the world does not have the necessary infrastructure and resources to tackle this pandemic.
Covid-19 pandemic requires technology-driven solutions for population screening, tracking the infection, prioritizing the use and allocation of resources,& designing targeted responses.
There is no way for countries to perform screening on all citizens as most of the countries do not have infrastructure and resources to do so. There are no solutions available for the identification and monitoring of the disease symptoms generation pattern.
We need solutions that can be utilized for a population of 7.8 billion. We need solutions that can monitor the health status of the entire country, trigger alerts for pandemic and help countries to utilize available resources in more precises way.
# Solution:
Vitality is a social accountability application that requires contribution from citizens and government officials. Using the vitality mobile applications citizens can register themselves and provides update related to their health daily, generate a request for necessary supplies, get latest updates from government officials.
Using the admin portal government officials can keep track of the user's health, identify the citizens having the highest probability of being a corona inflect person, provides necessary helps to the needy ones.
# Vitality Mobile Application:
Using the Vitality mobile applications citizens can register themselves
Provides update related to their health status to the government officials
Monitor their health status and progress
Perform self-assessment tests and get reports
Generate a request for necessary supplies and provide help to the needy ones
Get latest updates published by government officials
Check the inflected areas using the map
Maintain the social distance while moving out
# Vitality Admin Portal:
Using the admin portal government officials can keep track of the citizen’s health
Identify the citizens having the highest probability of being inflected person
Monitor and predict the pandemic outbreak
Provides the necessary help and support to the needy ones
Monitor the inflected citizens' progress
Notify the citizens for necessary measures and precautions.
# Outcome
The proposed solution will help the governments to utilize the available resources more efficiently.
Identification and prediction of pandemic diseases will improve.
The government will be able to monitor the citizens health in a much-planned way and provide necessary help timely.
Citizens will be able to monitor their health and perform their social responsibilities by helping the needy ones
# Learnings:
We leaned a new set of technologies for the creation of this application. We have learned flutter, nest.js, typescript and time management during this time.
What's next for Vitality: Your Social Accountability App
AI based Bots to perform a self-assessment test of the user
Option to chat with other users
Option to share experience during the lock-down
Creation of AI engine to predict the probability of inflection
Option to track the inflected person using the admin portal
Integration of blockchain for record keeping
Marketplace for medical supplies
Check Out The Progress
Vitality App
Vitality Portal
Built With
express.js
flutter
mongodb
nest.js
typescript
Try it out
drive.google.com | Vitality: Your Social Accountability App | Vitality is a social accountability application for population screening, tracking the infection, prioritising the use and allocation of resources,& designing targeted responses. | ['Saksham Chaurasia', 'vasa Saini', 'Ashish Kots', 'Parmesh Shiroya'] | [] | ['express.js', 'flutter', 'mongodb', 'nest.js', 'typescript'] | 7 |
9,900 | https://devpost.com/software/ewpi-early-warning-pandemic-indicator | (CoRST)
Inspiration
“Prevention is better than cure” for any virus to progress from endemic to epidemic to pandemic lifecycle stages like attachment, penetration, uncoating, biosynthesis, maturation, Time is of key essence. Whenever a virus enters host body, immune system gets activated and antibodies starts fighting against the virus which ultimately led to change in body vitals like Temperature, cough, shivering etc. These change with respect to each day or time in body vitals along with other health indicators like blood pressure, heart disease, lung disease, blood tests and diabetes becomes a vital clue to detect infection .
Our solution is to perform online screening for every possible individual using the chatbot on his/her system providing the details on frequent interval ( could be : every alternate day).
If symptoms for the effected Persons are getting reflected, our model will find the most probable person or area to be tested.
What it does
Takes all body vital parameters including sore, age, gender, wet cough, Tightness in your chest, Loss of taste and smell, Dizziness, confusion or vertigo, Headache, Muscle aches, Sore throat, runny or blocked nose, cold , low grade fever, temperature at defined interval of a day / time
Cough can be recorded via browser based microphone
Record breathe in and out from mouth as deeply as possible five times
Record read aloud any text “ with high pitch”
GPS location of the individual
Take rapid antibody test results data to check infection
Rule based Self-assessment to detect severity of infection
In case of individual not able to detect or unsure video call option with doctor to further analyse
To detect normal cough and Covid19 or similar other cough an speech model would be built
Data collected from GPS location would be used to detect hotspots of specific places to be quarantined
Since time is collected places nearby can also mapped
If the propagation of virus is human to human, airborne the same would trigger to nearby places that would trigger an alarm to authorities to seal off the complete area it can be plotted over google maps to check progression
Basis the factor provided by individual and time, Markov chain can describe the progression of virus lifecycle based on chronic and non-chronic diseased individuals
How I built it
Technologies and Frameworks we have used are Python, R, Angular, REST API, Scikit-Learn, MongoDB, Google Maps, NGINX Webserver, Tensorflow
Using the vital parameters, cough and GPs location, We intent to build an ensemble model for predicting the possibility of individual being infected with the virus.
With GPS location, identifying virus infected hotspots. If a location is not a hotspot, then based on increase in probable cases, we intend to predict the probability and time in which an area can become a hotspot using a markov model.
Challenges I ran into
Limited Cough data availability.
Lack of disciplined approach for providing input data to screening engine.
Unable to track Protein structure of Rapid Self Mutating Virus.
Accomplishments that I'm proud of
Accuracy of Model
What I learned
Limited Cough data availability.
Lack of disciplined approach for providing input data to screening engine.
Unable to track Protein structure of Rapid Self Mutating Virus.
What's next for Covid Risk Screening & Tracing (CoRST)
1)Based on the infected area, we can have multiple actions around food distribution and needy trackers in case of locked down situation.
2) Build the Analytics report based on the data collected
Built With
angular.js
ec2-servers-on-cloud
google-maps
mongodb
nginx
python
r
rest-api
scikit-learn
tensorflow | Covid Risk Screening & Tracing (CoRST) | CoRST will help people take online screening for Covid risk and storing screening summary on cloud, so that consecutive test results of an user can be compared to find risk for Pandemic. | ['Gaurav Srivastava', 'Sumit Sharma', 'Vishwas Kr Tomar', 'indu madan', 'Varun Anand', 'Mahendra s', 'Sanyam Jain'] | [] | ['angular.js', 'ec2-servers-on-cloud', 'google-maps', 'mongodb', 'nginx', 'python', 'r', 'rest-api', 'scikit-learn', 'tensorflow'] | 8 |
9,900 | https://devpost.com/software/itown | iTown is a free app for towns, villages and high streets that collates up to the second information from all the businesses in each local area that, during these unprecedented times, have had to adapt to the changing needs of the Covid-19 lockdown.
By creating a digital shop window, iTown enables consumers using the app to see what products are on offer locally in businesses such as corner shops, butchers, fruit and veg shops, bakers, and takeaway food outlets. They can then order and arrange delivery through a simple online messaging service and pay online.
Inspiration
What it does
How we built it
Challenges we ran into
Accomplishments that we're proud of
What we learned
What's next for iTown
Try it out
apps.apple.com | iTown | iTown is a ready-made app for towns, villages and high streets across the country in direct response to the current Coronavirus lockdown. | ['Victoria Mann', 'Martin Greenhalgh'] | [] | [] | 9 |
9,902 | https://devpost.com/software/lassio | Logo
Login Screen
Jobs Page
Labelling Page
Submission Received Page
User Profile
Introduction
What: Industrial data labelling-as-a-service
Why: Labelled data of sufficient quality is required to effectively utilize machine learning
How: Crowdsource energy industry’s data labelling jobs to independent subject matter experts
Who: Energy companies want their data labelled & industrial subject matter experts want to prove their expertise
Vision
Enabling and accelerating digital transformation of the energy industry by providing labelled data.
Machine learning and data labelling
Machine learning (ML) is already a billion dollar industry. Tens of thousands of people across the globe work with labelling data sets that machine learning algorithms can use for training. Labelling itself is a $150 million industry (2018). Heavy-asset industries, and the energy industry specifically, is lagging behind in their digital transformation journeys - applications of machine learning are limited to pilot projects and proof of concepts. One of the reasons that industrial data labelling has not yet been executed at scale has been a shortage of subject matter experts.
Crowdsourcing industrial data labelling
With hundreds of thousands of people losing or set to lose their jobs in the energy industry in the wake of the oil price collapse and COVID-19, there is a need for innovative business models that quickly can provide relevant work directly to highly-skilled energy professionals.
By crowdsourcing labelling of industrial sensor data, energy companies enhance their data sets - allowing for more accurate results in existing data-driven processes and enabling machine learning.
The World Economic Forum has estimated the value of digital transformation just in the Oil & Gas industry to $1.6 trillion. Labelled data is essential to get anywhere close to this number.
Solution
A web-based application where companies can post data labelling jobs, allowing subject matter experts around the world to:
Test their data labelling abilities on a data set. We compare their labelling accuracy to a small labelled data set provided by the customer
If they perform well on the test, we release the entire dataset to them, anonymizing company details to preserve confidentiality if necessary.
Upon completion of data labelling job, the labeller gets paid
The data labelling results are used to generate credentials on Hedera Hashgraph. The credentials objectively measure data analysis and labelling skills. These credentials can then be immediately verified by any other party, by leveraging blockchain technology.
The data is only released to the labellers qualifying for the job, limiting who gets access. Additionally, the company offering the labelling job may choose to keep itself anonymous
Built with
Java
Hedera Hashgraph / DID and Verifiable Credentials
React/JavaScript
Open source software (Gearbox.js, react-table, antd, Open Industrial Data)
Appendix
Presentation material:
Slides
Backend code:
Github
Frontend code:
Github
Figma Prototype:
Figma
Built With
antd
blockchain
hashgraph
hedera
iot
java
react
sensor-data
Try it out
github.com
github.com | Lassio | Enabling industrial machine learning by leveraging blockchain technology & crowdsourcing | ['Petter Reistad', 'Garrett Gillas'] | ['Future of Work 1st Place'] | ['antd', 'blockchain', 'hashgraph', 'hedera', 'iot', 'java', 'react', 'sensor-data'] | 0 |
9,902 | https://devpost.com/software/work-with-digital-approval | Uncle Jon: A.I powered Gatekeeper
As coronavirus recedes, companies will have a daunting challenge to choose which employees work from home and who come to the office. Is there a solution to this problem? Yes, Uncle Jon has an idea!
He suggests that each employee will have to take an A.I powered test which will evaluate how likely they are to be affected by COVID-19. Then, based on the results of the diagnosis test and the position of the employee in the company hierarchy, Uncle Jon will automatically suggest which employees should be called to the office and how they must be seated. These results will be shared with the HR team who can then modify it as per their needs.
Uncle Jon will issue a digital identity card to the employees who have to come to the office. The identity card would generate a decentralized token id, which the employee would have to enter at the office gate. The identity card would be valid for a period of 7 days, post which the employee will have to take the A.I based test again to validate their health.
The use of a decentralized token will help common people understand the power of blockchain and how it can be used to protect privacy and yet provide a simple platform for the safety of workplaces.
That’s Uncle Jon.
Main Idea: This test segregates to ensure high priority cases work from home, while low priority cases come at the office
Built With
html
javascript
machine-learning
Try it out
www.unclejon.xyz | Uncle Jon Gatekeeper - Your Digital Security Guard | A web-based automated security guard for acting as a gatekeeper for the company and their offices to monitor incoming visitors and check whether they are Coronavirus free. | ['Kavish Goel', 'Stuti Kalra', 'Taruna Garg'] | ['Future of Work 2nd Place'] | ['html', 'javascript', 'machine-learning'] | 1 |
9,902 | https://devpost.com/software/fuel-for-thought | Title Slide
User Homepage Wireframe
User Dashboard Wireframe
Home Page
Log In
Sign Up
Sample Assessment 1
Sample Assessment 2
Sample Assessment 3
Sample Assessment 4
Sample Assessment 5
Inspiration
In the midst of an economic downturn that severely affected the O&G industry, we were inspired to create a platform that keeps energy workers' minds sharp so that they're ready to return to work with minimal barriers for reentry.
What it does
Fuel for Thought is a multistep, ongoing process that keeps employees connected. Employees have their personality and skills assessed through situation-based exercises, which is securely stored on the Hedera Hashgraph for employers to view. Then, the platform recommends training modules to the employees for them to improve their current skills and learn new skills that are hot in the job market. These employees are then matched to maximize synergy with new and existing teams, creating units that large energy companies and service providers can quickly hire.
How we built it
We built Fuel for Thought by creating a website using HTML/CSS/jQuery/JavaScript, hosted on Flask. The site uses a custom-built REST API to access the Hedera Hashgraph for file storage and querying, as well as calling of the ML model that powers the team creation. We also used the p5.js to develop a demo game that assesses personality traits.
Challenges we ran into
One of the main problems we ran into was connecting the Hedera SDK with Flask. Since the Hedera SDK was built primarily to support JavaScript and Java, we had to convert the JavaScript code into a NodeJS script and then execute it in Python using Naked to sent the results to Flask.
Accomplishments that we're proud of
The team is extremely proud of being able to not only create a working website with demo games and dashboards, but also include a fully-fledge business-side presentation that presents the purpose and value of Fuel for Thought in an effective way.
What we learned
We learned the following:
p5.js
development of REST APIs
fundamentals of Node.js
Hedera Hashgraph and Hedera File Service API
What's next for Fuel for Thought
Development of a more robust dashboard and ML model that takes into account the vast amounts of novel data available to employers through the internet.
Built With
css
flask
hedera
html
javascript
p5
python
sklearn
Try it out
github.com
docs.google.com | Fuel for Thought | Fuel for Thought is an AI and hashgraph-based platform that allows energy workers to hone in on their technical skills while connecting them with other workers and large companies looking to hire. | ['Andrew Arkhipov', 'May Chen', 'Philip Kung', 'Murray Lee', 'Alexander Jeanis'] | ['Future of Work 3rd Place'] | ['css', 'flask', 'hedera', 'html', 'javascript', 'p5', 'python', 'sklearn'] | 2 |
9,902 | https://devpost.com/software/wellpass | Block Diagram
GIF
hospital registration
GIF
hospital verification by admin
GIF
patient requesting data from hospital after registration
GIF
hospital approving patient data
GIF
patient generating WellPass in from of QRcode
GIF
authorities verifying user on basis of his QRcode
Inspiration
In the post COVID-19 world there will be an urgent need to verify the medical records of the patient before he or she is allowed to travel around. The authorities will need a way to verify patient records for travelers. WellPass enables easy access of medical data to authorities like hotels/airports. The end users don’t need to carry their offline copies of their medical reports. The authorities just need a simple app or the web interface to check the validity of the claims by the users.
## What it does
WellPass is a web based application for securely storing and verifying patient data. We use Ethereum blockchain & IPFS (InterPlanetary File System) for secure and decentralized storage of medical records. The user requests his data from his dashboard. The approved hospital receives the patient’s request and adds the encrypted data to blockchain. After that a QRCode is generated for the user which stores the patient data. When the user arrives at the airport/hotel, he/she simply displays his/her code to the authorities, they in turn scan the code via the web application or the mobile application. If the data is valid they are allowed to work. This prevents any sort of forgery of data by anybody. We have also created an admin panel for authorizing the right hospitals.
How I built it
We have developed a full stack application using Nodejs,MongoDB & Express which uses JWT based authentication. The patient records are encrypted with Eth-crypto and then stored on ethereum blockchain via smart contracts & the reports are stored on the InterPlanetary File System(IPFS) network. QR scanner is implemented using JsQR library. The Android App uses Vision API from Google to scan the QR Code.
Challenges I ran into
Some challenges that we faced were creating an ethereum wallet for the users and the hospitals, using JWT authentication and implementing QR code scanner in our web app.
Using JWT authentication in NodeJs
Implementing Model-View-Controller pattern in backend
Setting MongoDB Atlas
Defining Complex Schema's in MongoDb
How to setup a full stack application using the MEN stack
How to Deploy a web app
Setting up express and using middleware for custom error handling
Accomplishments that I'm proud of
We successfully implemented an ethereum wallet that is stored in the user's browser. We also created a QR code from the blockchain data. Also making a production grade application which has a real world use case ## What I learned While working in this project, we learned how to implement ethereum blockchain in a web app. We learned how to use IPFS and store image data on it. The storage of patient data records required encryption so we used Eth-crypto to encrypt and decrypt the data. We also learned how to use Mongodb, along with Express, and Nodejs.
What's next for
Wellpass We plan to shift to some other blockchain with a different mining system so that the time a transaction takes to confirm can be reduced.
Built With
arcblock
ethereum
express.js
javascript
jwt
mongodb
node.js
solidity
visionapi
Try it out
thewellpass.herokuapp.com
github.com
github.com | Wellpass | Travel authorities need to verify the COVID-19 testing results of the travellers. WellPass is a web based application for securely storing and verifying patient data. | ['Sarthak Arora', 'Pranav Garg', 'Varun Ramnani', 'Akshit Suri'] | [] | ['arcblock', 'ethereum', 'express.js', 'javascript', 'jwt', 'mongodb', 'node.js', 'solidity', 'visionapi'] | 3 |
9,902 | https://devpost.com/software/predicting-job-imposters-01j4sm | Inspiration
COVID-19 pandemic is affecting economies in every continent. Unemployment rates are spiking every single day with the United States reporting around 26 million people applying for unemployment benefits, which is the highest recorded in its long history, millions have been furloughed in the United Kingdom, and thousands have been laid off around the world.
These desperate times provides a perfect opportunity for online scammers to take advantage of the desperation of thousands and millions of people looking out for jobs. We see a steep rise in these fake job postings where the posting seems genuine, often these companies will have a website as well followed by a recruitment process that is similar to other companies in the industry.
What it does
Our mission is to create awareness among the job seekers regarding the seriousness of this issue and how, through a machine learning model integrated in our designed app "Job", we can predict whether a job posting is fraudulent or not as well as job seekers can also apply for the particular listed job.
How I built it
We explored the data and provided insights into which industries are more affected and what are the critical red flags which can give away these fake postings. Then we applied machine learning models to predict how we can detect these counterfeit postings.
In further detail:
Data collection: We used an open source dataset that contained 17,880 job post details with 900 fraudulent ones.
Data visualisation: We visualised the data to understand if there were any key differences between real and fake job postings, such as if the number of words in fraud job postings was any lesser than real ones.
Data split: We then split the data into training and test sets.
Model Training: We trained various models such as Logistic regression, KNN, Random Forest etc. to see which model worked best for our data.
Model Evaluation: Using various classification parameters, we evaluated how well our models performed. For example, our Random Forest model had a roc_auc score of 0.76. We also evaluated how each model did in comparison to the others.
Immediate Impact
Police departments don’t have the resources to investigate these incidents, and it has to be a multi-million-dollar swindle before federal authorities get involved, so the scammers just keep getting away with it.
Hence our solution saves millions of dollars and hours of investigation, whilst protecting the workers from getting scammed into fake jobs and misused information.
Revenue generated
Our Revenue model is based on:
1) Premium subscription availability to job seekers to apply for jobs
2) Revenue from the advertisements
3) Commission from the employers to post the jobs
Funding Split
1) Testing and Development: $ 10,000
2) Team Hire Costs: $ 2000
3) Patent Application Costs: $ 125
4) Further Licensing conversations: $ 225
TOTAL: $ 12,350
Future Goals
We would hope to partner up with LinkedIn or other job portals in a license agreement, to be able to integrate our machine learning model as a feature on their portal.
Built With
adobexd
machine-learning
python
Try it out
xd.adobe.com
github.com
drive.google.com | Providing legit job postings by filtering job imposters | Spotting the differences between fake postings and genuine ones | ['Aerica Singla', 'Arushi Madan', 'Arun Venugopal'] | [] | ['adobexd', 'machine-learning', 'python'] | 4 |
9,902 | https://devpost.com/software/dappsule-democratizing-safer-workplace | Inspiration
..
What it does
..
How I built it
..
Challenges I ran into
..
Accomplishments that I'm proud of
..
What I learned
..
What's next for
Built With
blockchain
bootstrap
figma
kml
mongodb
php
qrtag
rfid
solidity
Try it out
www.dappsule.com | .. | .. | ['Praveen Kumar'] | [] | ['blockchain', 'bootstrap', 'figma', 'kml', 'mongodb', 'php', 'qrtag', 'rfid', 'solidity'] | 5 |
9,902 | https://devpost.com/software/care-at-home-juek57 | With Care@Home, Patients are able to attend better to their health needs in the comforts of their home. Patients used to wait and look for hours just to get a medical practitioner to take a look. They would jump from hospitals to another hospital just to find the care they need. Procedures or checkups they need seems to become an inconvenience rather than a need because of the current situation of mobility and access.
Now, with Care@Home and their integrated system, the patient journey has become easier and faster. Now, they can book a nurse at their convenience and schedule vital checkups and consultations without the hassle of leaving their homes.
Care@Home does an extensive screening of medical practitioners to ensure patients get the best healthcare possible.
They have also partnered with Insurance companies to ensure patients can access and process their insurance in a more convenient way.
They can rest easy knowing that they have access to doctors, nurses and other medical practitioners to care for their family at their fingertips.
Built With
django
invision
kotlin
python
Try it out
www.figma.com | Care at Home | Complete Healthcare at your Fingertips | ['Jann Alfred Quinto', 'Drea Juan', 'Einstein Rojas'] | [] | ['django', 'invision', 'kotlin', 'python'] | 6 |
9,902 | https://devpost.com/software/https-github-com-amir656-did-connect-examples | Inspiration
With no clear end in sight, the novel coronavirus has killed thousands, halted the economy and stolen countless experiences, jobs and memories. Contact tracing, or figuring out who was exposed to someone who tests positive, could greatly benefit from technological assistance as it's nearly impossible to remember everyone you came within 6 feet of in the past 2 weeks. Apple and Google have teamed up to create a system where your phone would broadcast a unique identifier to everyone within 6 feet of you. Your phone would simultaneously collect this list and would then check it against some list of confirmed covid-cases. The problem we aim to solve is providing a highly trusted and accurate list of covid-cases for these contact tracing solutions to utilize.
What it does
Apple and Google have teamed up to create a system where your phone would broadcast a unique identifier to everyone within 6 feet of you. Your phone would simultaneously collect this list and would then check it against some list of confirmed covid-cases. The problem we aim to solve is providing a highly trusted and accurate list of covid-cases for these contact tracing solutions to utilize.
Maintaining privacy is critical to ensuring wide-adoption. To that end, the contact tracing solution will likely generate random contact-tracing ID (CTIDs) every few hours to ensure that these CTIDs do not become a way to track someone's movements.
How we built it
Leveraging Epic's FHIR System (Fast Healthcare Interoperability Resources), we link covid-19 test results to all contact-tracing ids generated by a user in the 2 weeks prior to their positive test.
Utilizing ArcBlock's DiD, decentralized identifier, we provide a simple, secure way for users to enroll in contact tracing and connect their contact-tracing ids to their hospital
Challenges we ran into
Setting up the API's and conceptualizing the entire contact tracing process to discover where we could use these technologies.
Accomplishments that we're proud of
Integrating decentralized systems, front-end development and interfacing with a rich, multidimensional API to make real change
What we learned
The importance of documentation, and a ton of how the purposed contact tracing will be implemented.
What's next for
https://github.com/amir656/did-connect-examples
Scaling up to meet the demand and earn the trust of every American with a smart phone
Built With
arcblock
epic
fdhir
npm
react | https://github.com/amir656/did-connect-examples | covID: Integrating hospital verified test results to assist in contact tracing to help beat Covid-19 | ['Rami Shahatit', 'Amir Shahatit'] | [] | ['arcblock', 'epic', 'fdhir', 'npm', 'react'] | 7 |
9,902 | https://devpost.com/software/pneumoscan-an-ai-radiology-tool-for-covid-19-pandemics | Fig. 1: Map of Covid19 cases around the world (as of 4/30/2020)
Fig 2: Top 10 countries with most COVID-19 deaths
Fig 3: Current chest X-ray diagnosis vs. noval process with CovidScan.ai
Chart of wait-time reduction of AI radiology tool (data from a simulation stud reported in Mauro et al., 2019).
Fig. 5: Process of CovidScan development
Demo of web-app:
https://www.cv19scan.site/
(Please use Internet Explorer, or Firefox, our web-app currently doesn't support Chrome)
Dataset:
For the data analytics of COVID-19 pandemics, we used data collected by the Johns Hopkins University Center for Systems Science and Engineering updated on 4/30/2020.
For the chest X-ray detection models, we used combined 2 sources of dataset:
The first source is the RSNA Pneumonia Detection Challenge dataset available on
Kaggle
contains several deidentified CXRs with 2 class labels of pneumonia and normal.
The COVID-19 image data collection repository on
GitHub
is a growing collection of deidentified CXRs from COVID-19 cases internationally. The data is collected by Joseph Paul Cohen and his fellow collaborators at the University of Montreal
Eventually, our dataset consists of 5433 training data points, 624 validation data points and 16 test data points.
Inspiration
What will be working situation for medical staff in hospitals during and after the COVID-19 pandemic? How can the medical staff quickly and securely log in and perform PPE safety check while dealing with a huge influx of patients in critical conditions? How can we automate the process of COVID-19 diagnosis so precious time can be saved for both medical doctors and the patients? How can our solution for hospital later be scaled and implemented to be a essential tool for automating the daily operation at hospital even after the COVID-19 pandemics is over?
To answer these core questions, we did some background research to identify the main challenges in order to develop the best solutions around those:
COVID-19 Pandemic:
Fig. 1: Map of Covid19 cases around the world (as of 4/30/2020). Our team created the map based on data collected by the Johns Hopkins University Center for Systems Science and Engineering.
As we see from the map above and the pie chart below, COVID-19, previously known as the novel Coronavirus, has killed more than 63,860 people and infected over 1,067,061 people in the United States alone, topping all other countries around the world. This number is continuing to grow every day.
Fig. 2: Top 10 countries with most COVID-19 deaths.
The 3 main problems occur in the healthcare system during the pandemics are:
1. Confidentiality:
As you may see on the news, hospitals all over the U.S. (New York, Chicago,California…) and other countries (Italy, Spain…) are flooded with a huge influx of patients with critical conditions. With the increasing workload for the medical staff, patients’ confidential information may be put at risk if unauthorized personels can hack into the electronic medical record system. Thus, there is a need for a fast and secured method for medical staff to log in to the electronic medical record platform, so that the staff can move quickly with patients’ information inputting and still remain compliant with HIPPAA (Health Insurance Portability and Accountability Act). Badge scanning will be highly secured solution for this problem.
2. PPE Safety Check:
According to CDC, during COVID-19 pandemics, all healthcare workers should follow strict guidlines and protocols from OSHA regarding wearing PPE. All of the PPE prevents contact with the infectious agent, or body fluid that may contain the infectious agent, by creating a barrier between the worker and the infectious material. Gloves, protect the hands, gowns or aprons protect the skin and/or clothing, masks and respirators protect the mouth and nose, goggles protect the eyes, and face shields protect the entire face. N95 masks are the PPE most often used to control exposures to infections transmitted via the airborne route. Therefore, checking medical staff’s PPE safety protocol is especially crucial during this pandemics.
3. Long wait time for COVID-19 chest X-ray result:
Fig 3: Current chest X-ray diagnosis vs. novel process with CovidScan.ai
Patients can first be screened for flu-like symptoms using nasal swap to confirm their COVID-19 status. After 14 days of quarantine for confirmed cases, the hospital draws the patient’s blood and takes the patient’s chest X-ray. Chest X-ray is a golden standard for physicians and radiologists to check for the infection caused by the virus. An x-ray imaging will allow your doctor to see your lungs, heart and blood vessels to help determine if you have pneumonia. When interpreting the x-ray, the radiologist will look for white spots in the lungs (called infiltrates) that identify an infection. This exam, together with other vital signs such as temperature, or flu-like symptoms, will also help doctors determine whether a patient is infected with COVID-19 or other pneumonia-related diseases. The standard procedure of pneumonia diagnosis involves a radiologist reviewing chest x-ray images and send the result report to a patient’s primary care physician (PCP), who then will discuss the results with the patient.
_Fig 4: Chart of wait-time reduction of AI radiology tool (data from a simulation stud reported in Mauro et al., 2019). _
A survey by the University of Michigan shows that patients usually expect the result came back after 2-3 days a chest X-ray test for pneumonia. (Crist, 2017) However, the average wait time for the patients is 11 days (2 weeks). This long delay happens because radiologists usually need at least 20 minutes to review the X-ray while the number of images keeps stacking up after each operation day of the clinic. New research has found that an artificial intelligence (AI) radiology platform such as our CovidScan.ai can dramatically reduce the patient’s wait time significantly, cutting the average delay from 11 days to less than 3 days for abnormal radiographs with critical findings. (Mauro et al., 2019) With this wait-tine reduction, patients I critical cases will receive their results faster, and receive appropriate care sooner.
What it does
Using the power of pretrained machine learning models from open source, CovidScan.ai is created as a full-scaled AI tool for radiology clinics and hospitals. It can automate the process of security log-in, PPE safety check for medical staff and assist radiologists determine sign of COVID-19 on chest X-ray images with high accuracy indicates pneumonia. This tool of cutting edge technology can be used to reduce the workload for clinicians, and speed up patients’ wait time for pneumonia lab results in this critical time of the COVID-19 pandemic.
Fig 5: Deployment process of pretrained ML model to the web-app
As explained in the figure above, the CovidScan web-app includes 3 main AI components:
1. ID Badge Scanner:
For security purpose, only authorized personel can access to the web-app, which contains patients’ confidential health information (name, date of birth, chest X-ray, medical history…). Hence, the web-app will use pretrained scan the medical’s badge to grant them access to the software.
2. PPE Safety Check:
Due to hospitals/clinics’ strict guidelines in PPE usage, especially during this COVID-19 ourbreak, the web-app will ask the medical staff if he/she is in direct contact with patients for chest X-ray taking. If yes, then the web-app witll use AWS pretrained to check for medical staff’s PPE to see if the staff follow the safety protocols to minimize any exposures to the disease. If the medical staff passed both the secured check and safety, he/she can move on the the next step.
3. COVID-19 Chest X-ray Testing:
In the last step, the medical staff take patients’ chest X-ray images using the specialized machine and then upload the taken images to the database of web-app for testing for sign of COVID-19 infection or bacterial pneumonia.
It is due to the fact that an AI system can review, highlight the pneumonia sign and classify each X-ray image all in less than 10 seconds (comparing the radiologist’s 20 minutes that we mentioned earlier), and it can do that same task effortlessly for 24 hours without taking a break. This time cut is especially critical in the time amid the pandemic of COVID-19. With this spreading rate, it will be overwhelming for radiologists to review a massive number of chest X-ray images of potential COVID-19 infected patients. With the assistance of CovidScan.ai, it can automatically highlight the suspected signs of pneumonia for the radiologists and speed up the process of chest X-ray review. Therefore, more COVID-19 positive-tested patients will get their result back faster and receive appropriate care sooner to prevent the spread of the virus.
How we built it
Employee Badge Scanner:
We developed this feature using the open-source python library Pyzbar. We have written the script in the JQuery which sends the snapshots from the live camera feed to the inference model at the backend. It can read one-dimensional barcodes and QR codes present on the employee’s ID badge. We implemented this feature to work with a snapshot of employees’ ID badge.
Link:
https://pypi.org/project/pyzbar/
PPE Safety Check:
We developed this feature using the open-source TensorFlow model for face mask detection. The backbone network only has 8 Conv layers and the total model has only 24 layers with the location and classification layers counted. The dataset is composed of WIDER Face and MAFA datasets. We have written the script in the JQuery which sends the snapshots from the live camera feed to the inference model at the backend. It works with live footage from any sort of cameras and detects people not wearing a face mask.
Link:
https://github.com/AIZOOTech/FaceMaskDetection
Chest X-ray Classification:
For this feature, we developed a Pytorch model. This project’s goal is to draw class activation heatmaps on suspected signs of pneumonia and then classify chest x-ray images as “Pneumonia” or “Normal”. For this project, we are going to use a dataset available at Kaggle consisting of 5433 training data points, 624 validation data points and 16 test data points. C. For the model, we load the pre-trained Resnet-152 available from Torchvision for transfer learning. ResNet-152 provides the state-of-art feature extraction since it is trained on a big dataset of ImageNet. ResNet-152, as the name sounds, consists of 152 convolutional layers. Due to its very deep network, the layers are arranged in a series of Residual blocks. These Residual blocks skip connections to help prevent the vanishing gradients which are a common problem with networks with deep architecture like ours. Resnet also supports Global Average Pooling Layer which is essential for our attention layer later on. For the attention layer to draw the heatmap, we use the global average pooling layer proposed in Zhou et al. Global average pooling layer explicitly enables the convolutional neural network (CNN) to have remarkable localization ability. We achieve 97% accuracy on the training dataset and 80% on the testing dataset.
Web development: The trained weights of the deep learning models are deployed in a form of Django backend web app CovidScan.ai. While the minimal front-end of this web app is done using HTML, CSS, Jquery, Bootstrap. In our latter stage, the web-app will then be deployed and hosted on Debian server.
Technical Requirements:
The packages required for this project are as follows:
Torch (torch.nn, torch.optim, torchvision, torchvision.transforms)
Django
Numpy
Matplotlib
Scipy
PIL
Tensorflow
jQuery
Challenges we ran into
This hackathon project was a very different experience for us which challenged us throughout this project with the AWS sagemaker. This is the first time we all were working with AWS sagemaker and creating endpoints of the pre-trained TensorFlow model. Also, understanding curated models and determining their accuracy was a little bit challenging for us. Even after successfully deploying the model’s endpoints, calling Amazon SageMaker model endpoints using Amazon API Gateway and AWS Lambda gave us a very hard time.
Accomplishments that we're proud of
We manage to finish the project in such a limited time of 2 weeks in our free time from school and work. We still keep striving to submit on time while learning and developing at the same time. We are really satisfied and proud of our final product for the hackathon.
What we learned
Through this project, we learn to implement a complicated image-recognition deep learning models from AWS marketplace. We also learn the process of developing a mini data science project from finding dataset to training the deep learning model and finally deploy & integrate it into a web-app. This project can’t be done without the efforts and collaboration from a team with such diverse backgrounds in technical skills.
What's next for CovidScan:
In the next 2 months, our plan is:
We will raise fund to invest more into the R&D process.
We will partner with research lab to collect more dataset and find hospitals to test our solution. One of our memeber has published his newly collected dataset on this open-source github:
https://github.com/nihalnihalani/COVID19-Detection-using-X-ray-images-/
Regarding our R&D, we plan on improving the performance of the platform, preferably by reading more scientific literature on state-of-art deep learning models implemented for radiology.
We also plan to add the bound box around the suspected area of infection on top of the heatmap to make the output image more interpretable for the radiologists. We are working to implament the multilabeling model of COVID-CXR on our dataset to improve our application. This model is published by The Artificial Intelligence Research and Innovation Lab at the City of London's Information Technology Services division and has accuracy 0.92, precision 0.5, recall 0.875, auc 0.96.
In many pieces of literature, they mentioned developing the NLP model on radiology report with other structured variables such as age, race, gender, temperature... and integrating it with the computer vision model of chest X-ray to give the expert radiologist’s level of diagnosis. (Irvin et al., 2019; Mauro et al., 2019) We may try to implement that as we move further with the project in the future.
With the improved results, we will publish these findings and methodologies in a user-interface journal so that it can be reviewed by expert computer scientists and radiologists in the field.
Eventually, we will expand our classes to include more pneumonia-related diseases such as atelectasis, cardiomegaly, effusion, infiltration, etc. so that this platform can be widely used by the radiologists for general diagnosis even after the COVID-19 pandemics is over. Our end goal is to make this tool a scalable that can be used in all the radiology clinic across the globe, even in the rural area with limited access to the internet like those in Southeast Asia or Africa.
References:
Crist, C. (2017, November 30). Radiologists want patients to get test results faster. Retrieved from
https://www.reuters.com/article/us-radiology-results-timeliness/radiologists-want-patients-to-get-test-results-faster-idUSKBN1DH2R6
Irvin, Jeremy & Rajpurkar, Pranav & Ko, Michael & Yu, Yifan & Ciurea-Ilcus, Silviana & Chute, Chris & Marklund, Henrik & Haghgoo, Behzad & Ball, Robyn & Shpanskaya, Katie & Seekins, Jayne & Mong, David & Halabi, Safwan & Sandberg, Jesse & Jones, Ricky & Larson, David & Langlotz, Curtis & Patel, Bhavik & Lungren, Matthew & Ng, Andrew. (2019). CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison.
Kent, J. (2019, September 30). Artificial Intelligence System Analyzes Chest X-Rays in 10 Seconds. Retrieved from
https://healthitanalytics.com/news/artificial-intelligence-system-analyzes-chest-x-rays-in-10-seconds
Lambert, J. (2020, March 11). What WHO calling the coronavirus outbreak a pandemic means. Retrieved from
https://www.sciencenews.org/article/coronavirus-outbreak-who-pandemic
Mauro Annarumma, Samuel J. Withey, Robert J. Bakewell, Emanuele Pesce, Vicky Goh, Giovanni Montana. (2019). Automated Triaging of Adult Chest Radiographs with Deep Artificial Neural Networks. Radiology; 180921 DOI: 10.1148/radiol.2018180921
Wang, L., & Wong, A. (2020, March 30). COVID-Net: A Tailored Deep Convolutional Neural Network Design for Detection of COVID-19 Cases from Chest Radiography Images. Retrieved from
https://arxiv.org/abs/2003.09871
Built With
matplotlib
numpy
pil
pytorch1.0.1
torchvision0.2.2
Try it out
gitlab.com
www.cv19scan.site | CovidScan-An AI Radiology Tool For COVID-19 Pandemic | CovidScan.ai is developed to be a secured AI platform with the purpose to assist radiologists with fast and accurate pneumonia dectection amid this COVID-19 pandemic. | ['Moksh Nirvaan', 'Nihal Nihalani', 'Vi Ly'] | ['Second Place', '2nd Place - Website Feature'] | ['matplotlib', 'numpy', 'pil', 'pytorch1.0.1', 'torchvision0.2.2'] | 8 |
9,902 | https://devpost.com/software/get-work | Trademark logo
Inspiration Booking and offering
Services in accordance with 21st century.
What it does
Imagine If your car is broken down in the middle of the way and you are having an app which assures you that a mechanic is on the way to fix it!
How I built it
My team will built it in next three months
Challenges I ran into
Innsuficiant fundings, expensive patent application
Accomplishments that I'm proud of
Lot of people and investors very interested in this projects to use a app
What I learned
I learn that market deperatly need this app because coronavirus
What's next for Get Work®
We are going to ekspand over boundries in one year after building app. We are in the process of securing funding of aprox. 50.000 EUR trough eather FIL ROUGE CAPITAL
which is platform for investors in Croatia or trough HAMAG BICRO credit line for entrepreneurs in
Croatia.
Which vertical do you consider the best match for your venture project? *
Digital Industry
Team composition and fit of each member with the project: *
Describe the core knowledge and expertise of each team member, as well any areas of competence
still in need of development within the team. Be as open and honest as possible. Describe also the
team's access to knowledge and expertise in any specialist field outside yours but required to
implement the solution envisaged, i.e. your product and / or service, and to establish your venture.
The founder of startup is Mario Marević. He studied Economy at "Andrija Kacic Miosic" High school in
Makarska and later on enrolled University of Split, Faculty of Economics in Makarska (2015-2018)
where he earned the title Bachelor of Commerce and Tourism Economy.
For the last couple of years, he has held high management positions at Albatros Real Estate, Adriatic
Homes Real Estate and at Makarska IN Travel Agency. He also worked as the seller of concession
services for Komunalno Makarska d.o.o., and as Sales representative for Jamnica d.o.o. (2014.) and
Studenac d.o.o. (2013.). Working on above mentioned positions he acquired communication,
organizational, management and marketing skills required for running his own company. He took
part in international project “Active Youth Entrepreneurship Network“ (AYEN) where he gained
valuable knowledge needed for running a business. Mario is a reliable, principled and responsible
person with strong organisational skills, able to perform tasks promptly and take responsibility for
the business, but at the same time to function as part of a team.
The rest of the team is made of three developers – students of Computer Science at University of
Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Split (Croatia).
Petra Dundić have a great command of programming languages VB, C, C++, HTML, CSS, data
structures, web tehnologies, algorithms, database systems, math for computer science, electronics
and object oriented programming.
Ivona Knaus have great command of programming languages VB, C, HTML, CSS and good command
in HTML, CSS , JS and data structures as well as OO languages (C++, C#, Java, Phyton).
Ante Sorić have a great command of c,c++, c# and java language. He has knoweledge in Digital
technology, Digital Computer Architecture, Algorithms and data structures, Database, Programming
for the Internet, UNIX Programming, Object oriented programming, Operating systems and
Computer and data security.
The team has all necessary knoweledge and experiences needed to sucessfully realize the project.
Addressed market/business pain and its dimension: *
Through a 2 to 3 sentence statement, describe who is your customer and the problem that your
product and/ or service will actually solve or help solving. Characterise clearly and quantitatively the
current problems faced by customers with existing solutions (from competitors, direct and indirect).
Market segmentation:
Geographic:
•Urban consumers (Split, Zagreb, Rijeka, Osijek)
Demographic:
•Working age (15-64 years)
Behavior:
•Service seekers — anyone who wants to book a service
•Service providers — Freelancers (choose this style of work because they enjoy the flexibility); Casual
earners (supplementing their income from a traditional job); Reluctant (take on-demand work out of
necessity but would prefer traditional job); The cash-strapped (employed in low-paid jobs, but have
to take extra work to make ends meet); Students
SERVICE SEEKERS PROBLEM:Finding the right service provider is time-consuming because the
information needed for decision-making are scattered in different places or not available at all
People do much more research before making a purchase decision and factors that are taken in
consideration before purchasing a service are:
• QUALITY — Customers attach great importance to recommendations, reviews and
experience of service provider.
• DELIVERY TIME — The era of technology has ushered in a practice of instant gratification of
demands why waiting for a service has become a story of the past. Customers want to know the
particular time to get their delivery and a given time duration.
• LOCATION In order to execute traditional services, people need meet in person, so they
prefer services within their reach to access it speedily.
• PRICE — Customers want to compare prices before making a purchase.
SERVICE PROVIDERS PROBLEM: It is almost impossible to legally acquire more income opportunities
(either stable or side income) without huge investments in marketing or opening a company.
Technology underlying the product or service; uniqueness of the solution: *
Please, keep in mind that the Venture Program targets deep-tech based solutions. Start by describing
your technology, the stage of technical development you are at and what the next critical steps in
development are to establish your MVP and have your solutions ready to launch in the market. What
makes it different (better) from existing solutions? Participants must refrain from disclosing any
confidential information for that purpose. In other words, do not describe how your technology
works but rather explain what technology it is, what it does and how it meets customers’ needs. Also,
explain your strategy for managing any Intellectual Property (IP) protection for your solution.
GET WORK is an mediator app between service providers and service seekers.
It is the only app where
one can find reviews, price, location, experience and contact of service provider in one place,
focus is on C2C type of mediation and
on traditional services where 2 people phisically meet like:
• Home care and design (interior designing, pest control, cleaning, laundry, glasswork,
woodwork, waterproofing, masonry, carpentry...)
• Repair and maintenance (handyman, locksmith, plumber, electrician...)
• Health, wellness, and beauty (manicure, pedicure, physiotherapy, massage, trainer...)
• Others (nanny, tutor, cooking, walking pets, shooping...)
This solution is disrupting the traditional business models, replacing them with a platform to connect
the major stakeholders of a service exchange (Service provider and customer).
Current MVP is a web page where one can offer or book services. The process is not automated yet
so applications, bookings and all communication is first going to administrator. Furthermore, the
implementation of Google maps is missing so locations are added manually on the map. There is no
possibility to create and edit a profile or to leave reviews and comments. There is currently 14 service
providers offering their services trough the web page. There are 60 interested service seekers that
are interested in trying on the app once it is made.
Future app functionalities:
• service provider info (name, photo, education, professional experience, list of services)
• service provider (and seeker) rating, reviews and comments
• posibility for service seeker to determine suitable time and place where he wants the
service to be provided — once provider agrees this way they both know particular time
of delivery and a time duration
• push notifications (when service is booked, when booking is accepted, when
appointment time is changed, reminder to leave a rewiew)
• location of provider on map + real-time tracking
• price of services provided (HRK/h)
• browsing trough options with filters like service type, location, price and reviews
• recommendations of service based on proximity, reviews, comments and price.
• quick and easy payment mode (cash or card)
• provider contact info — available once the service is booked which allows easy
communication
• overview of all available service options, while the most quality ones will appear on top
There is registered patent for app and name of app is registered trademark.
Name of patent: SUSTAV ZA POVEZIVANJE PONUDE I POTRAŽNJE USLUGA, reported at Croatian
Herald of Intellectual Property (Hrvatskom glasniku intelektualnog vlasništva) No. 5/2019 (8th of
March 2019) Patent No. P20171297A.
Holder of intellectual property rights: Mario Marević, Podgorske skale 9, Makarska
Product/service innovativeness: *
Through a clear, 2 to 3 sentence statement, describe your solution, the benefits for the customer and
how it overcomes the problems identified in item 2. Be as quantitative as possible in the description
of the benefits. For instance, do not just say that your solution “is better than..” but rather “enables
XX% improvement in..”
App solves SERVICE SEEKERS PROBLEMS because it provides them with all relevant information about
all available services in one place which makes the process of browsing trough, comparing and
booking offers fast, easy and convenient. Booking a service is possible in 30 seconds!
App solves SERVICE SEEKERS PROBLEMS because it enables them to start offering their services and
make an (extra) income with low initial investment in terms of time, knowledge, and money. It
enables business promotion, visibility of services, digitalization of business and therefore more
business opportunities. At last, it allows them flexible working hours.
Global impact: *
Describe the main characteristics (margins; emerging; consolidation stage, other) of the market in
which you will introduce your solution (product or service) and in which way your value proposition
(to the customer) makes it an attractive opportunity for an investor. Include the financial needs for
implementation and the projected returns for a hypothetical investor. Also, highlight how the market
has been growing and how it is expected to grow in the foreseeable future. Always quote and
reference your sources. You will also need to focus on market sizes for the total and the addressable
markets respectively, and make clear any economic, political, regulatory issues that may limit market
access.
On demand Market size and trends: According to Harvard Business every year on-demand economy
attracts over 22.4 million consumers for spending $57.6 billion. The CBInsight report revealed that 23
out of 310 private companies who valued at $1 billion by Jan 2019 belong to On-demand industry.
On-demand labor market is expected to flourish by 18.5% annually in the next 5 years and according
to the PwC report and will reach a value of $335 Billion by 2025. Global Online On-demand Home
Services market is expected to grow at a “CAGR of 53% with revenue USD 1,574.86 billion” by 2020-
The "YOY growth rate for 2020 is estimated at 32.14%" by the end of 2024.
By 2025. 72% of internet users will access the web solely via smartphones and 69% of 18-39
definitely use mobile devices to research products. In the first 5 years target market is Android users
of working age population (15-64 years old) of 4 major Croatian cities. Croatia has 4,388,476 mobile
subscribers in total and a 102,42% penetration rate. (4Q 2018) Market share of mobile vendors in
Croatia from Mar 2019 - Mar 2020 revieled tar Android has 78.78%.
In Split there are 121.242 people who are 15 to 64 yo, among them 95.781 Android users. In Zagreb
there is 517.212 working age population or 408.597 Android users. In Rijeka there is 88.544 people
age 15 to 64 yo and 69.950 Android users. In Osijek there are 73. 919 people 15 to 64 yo and 58.396
Android users.
GROWTH OF INTEREST AMONG SERVICE PROVIDERS
Unemployment rate goes up. There was 8.995 unemployed people in Split, 16.548 in
Zagreb, 3.700 in Rijeka and 5.049 in Osijek in March 2020. There was 8,3% (152.590)
unimployed people in Croatia. Unemployment Rate in Croatia averaged 16.9% from 1996
-2020. According to Trading Economics unemployment Rate in Croatia is expected to be
around 13.50% in 2021.
By 2020, the number of freelancers is projected to outpace full time workers - 73% are
looking for a job on dedicated internet platforms.
Because of economic crisis more people will need an extra source of income —79% of
existing users in on-demand economy are working as part-time.
GROWTH OF INTEREST AMONG SERVICE SEEKERS
Shortage of manual workers because od a) migration of labor abroad and b) higher level
of education of a growing part of the population.
Increased demand for blue collar service because of a) rising living standard and b) lack
of knowledge, tools or time to do small home repair and chores among young urban
population.
Financial means needed are for covering first year of operational expenses:
HRK EUR
Computers 24.000 3.150
Marketing 40.000 5.249
Office rental 18.000 2.362
Bookkeeping 12.000 1.575
Sallary 269.000 35.302
Total 363.000 47.638
Risks associated with your business model: *
Global investors expect teams responsible for their projects to apply international best practice in
managing the risks and achieving the objectives of the project. Risk management needs be balanced
with the opportunities presented in a timely and cost effective manner. Clearly identify the main
risks involved in your business model and strategy. Document the risks and the severity of any
legal/regulatory issue (e.g. International standards, issues with privacy regulation, IP disclosures and
protection strategies, etc.). Provide a discussion of the budgetary implications of dealing with the
above risks being honest about possible costs no matter how uncertain such estimates may be.
The relevant risks of this project and the introduction of an innovative service to the market are
outlined below:
Risk - Likelihood - Impact
APP FUNCTIONALITY, TECHNOLOGY AND DESIGN FAILURES – Medium Likelihood - High Impact
Mitigation or risk avoidance measures
work with reputable programmers on the app development
agile approach and frequent testing with customers
DEVELOPMENT TAKES LONGER THAN ANTICIPATED- Medium Likelihood – Medium Impact
Mitigation or risk avoidance measures
Consult with team regarding the service launch procedure
Define implementation timeline
Engage additional human and material resources to achieve the timetable
Harmonize the project timetable without jeopardizing the project results
INSUFFICIENT NUMBER OF APPLICATION USERS – Low Likelihood - High Impact
Mitigation or risk avoidance measures
Present new opportunities to meet the requirements of existing and potential clients
Follow the business plan and marketing strategy
Ensure project visibility
Personal contact to potentional service seekers and providers
Intensify potential customer visits to identify reasons
Present company offerings in other potential markets
Revise marketing plan
Correct customer offers (eg discounts, additional benefits)
LEGAL REGULATIONS (GDPR, PAYMENTS AND TAXES) Medium Likelihood – High impact
Mitigation or risk avoidance measures
consulting with lawyers and tax advisors prior starting project or entering new market
apply for and ensure all the required norms are satysfyed and certificates obtained to avoid
any legal issues
continue regular consultations with experts and follow up on changes in legislation
adjust functionality to legal regulations
LOW QUALITY OF SERVICE PROVIDERS – Low Likelihood – Medium Impact
Mitigation or risk avoidance measures
background checks on the people that work through the app
rating and comments of app users
creating an algorithm which does not reccomend service providers with low rating and bad
comments
LACK OF THE REQUIRED PROFILE PERSONELL TO WORK ON AN INNOVATIVE SERVICE Low
Likelihood - Medium Impact
Make a quality call for applications and a detailed job description
Offer a salary that meets the needs of the required profile
creating an application administration manual that will shorten the learning and adjustment
time of new employees
Repeat a job offer
Present a job at a local employment office
LACK OF FINANCIAL RESOURCES - Low Likelihood - High Impact
Mitigation or risk avoidance measures
Develop a quality project proposal
Apply for and implement various alternative financing options for the project (banks, EU
grants, National grants, Investors, Crowdfunding)
PROJECT COSTS ARE NOT WITHIN THE PLANNED BUDGET- Low Likelihood – High Impact
Mitigation or risk avoidance measures
Develop a budget based on real market prices that is in line with acceptable costs
Determine real market prices when designing tender documents for the procurement of
equipment and services
Find additional financing options if budget exceeds total planned amount
Reallocate funds in the budget
Activate additional financing options if the budget exceeds the total planned amount
Built With
c
c#
c+
c++
java
javascript
Try it out
drive.google.com | Get Work® | Get Work® - Mediator app between service providers and service seekers. Booking and offering services in accordance with 21st century. | ['Mario Marevic'] | [] | ['c', 'c#', 'c+', 'c++', 'java', 'javascript'] | 9 |
9,902 | https://devpost.com/software/research-platform-9fvt73 | Covid-19
The project is being developed in cooperation with the Crisis_Managment_Plattform
This project is a new type of research and an unprecedented global project that connects scientists, researchers and inventors in a way that has never existed before.
The data, which is crucial for research, is compiled internationally and is available to all researchers, scientists and inventors of this platform.
However, this is not just an international scientific project, it also includes a sociological project. It is a triple scientific project that has a great future. Ask yourself the question of what progress we can make in all areas of research if, instead of 200 scientists, 50,000 scientists, researchers and inventors research the projects together. With an incredible range of knowledge and ideas Thum bundled on a platform for the benefit of all people. The planning and organizational measures result from the association of international scientists, researchers and inventors. With the help of this platform, they organize virtual meetings to conduct research. Data is analyzed by a team of experts and integrated into the platform, so that everyone on this platform is always up to date with the latest research. Users of this platform also have the opportunity To propose a certain page, for example a research institute, to our team and after an examination by a team and consultation with the operator of the page, this is intrigued, and research institutions can have your page integrated so that all users on the platform see the information on the page The good thing about this platform is that users can organize themselves, plan meetings and exchange information at any time. At the same time, users have the opportunity to publish their discoveries internationally with all users of the platform.
Who is actually behind this whole campaign?
My name is Michael Rhein, I managed the patent exploitation TIZ-NORD Wilhelmshaven Technical Innovation Center for Research and Patent Exploitation. When the pandemic started, I thought about the options for dealing with this crisis as quickly as possible. So in the beginning I looked for solutions on my own until I noticed that the solution can be found right here. Because of this, I thought about how I could bring as many people as possible together to find a solution together. However, it was clear to me from the beginning that this would only be possible with experts in the fields of science, research and the inventions or medical means of the inventors resulting therefrom and so I developed the project of the research platform. I was aware that this would and will present me with enormous challenges, but I will be able to master these together with everyone.
What about the rights?
I have long considered how to organize the legal part of this platform. Now I have found a solution for this too. All those who register on this platform simply agree that the rights to the research results made on the research platform, inventions for the benefit of mankind, to all users registered on the platform who are involved in the creation of this invention, solution or the research results are evenly distributed . This is an optimal solution for doing research together. To implement this regulation, users accept the general terms and conditions of the platform, in which all legal points are defined.
What does it look like financially? Who finances the whole thing?
This platform is not about making the greatest possible profit, rather it is about developing things together that can avoid such a situation as Coovid-19 or at least bring about a quick solution. I try to get political funding from the governments from which the scientists, inventors and researchers come, which are then distributed to the respective research institutes, scientists, researchers and inventors, for this I will put together a finance team. We are all human beings, we all live on this planet, let's develop technologies and opportunities together that we may not even be able to imagine today. This platform, like all of us, lives from constant change and adaptation. Together we all develop this platform, so we become part of it.
Built With
api
audio
google-analytics
google-translate
html5
java
javascript
php
video
Try it out
forschungsplattform.com
forschungsplattform.com | Research Platform | This project is a new type of research and an unprecedented global project that connects scientists, researchers and inventors in a way that has never existed before. | ['Michael Rhein'] | [] | ['api', 'audio', 'google-analytics', 'google-translate', 'html5', 'java', 'javascript', 'php', 'video'] | 10 |
9,902 | https://devpost.com/software/weacademy | VIRTUAL CO WORKING COMMUNITY MARKETPLACE WITH ONE CURRENCY - Time
A SaaS monthly subscribtion platform where Anyone. Everyone. Anywhere. can post what he needs or offer what he/she can contribute back with within their expertise. our AI will measure each individual task och the value your time by experience, market and various measurement algorithms to determine who will be matched with each other.
YOU NEED A LOGO VS I NEED PITCHDECK = WeAcademy, the WE family
We are a team of experienced tech VC backed startups, we would love to show you a short demo of this amazing new way of accelerate building your MVP or helping established startups to grow, detect flaws in processess or gain more profit!
We need you to help us invest to make us complete our AI and go2market!
weacademy.co
Built With
commnunity
saas
Try it out
www.weacademy.co | WeAcademy | AI driven Exchange Assignments & tasks. Our only currency will be Time. | ['Beidos Ali'] | [] | ['commnunity', 'saas'] | 11 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.