README.md

# Web Scraping, Processing, and Embedding
This Google Colab notebook demonstrates a complete workflow for extracting information from a website, processing it, and preparing it for use in applications like question answering or semantic search.
Here's a breakdown of its functionality:
1. **Web Scraping:** It starts by crawling a specified website (in this case, `learn.microsoft.com/en-us/`) and extracting the text content from multiple pages.
2. **Data Processing:** The extracted text is then processed by splitting it into smaller, overlapping chunks. This makes the text more manageable for further analysis and embedding.
3. **Embedding:** Using a pre-trained Sentence Transformer model, each text chunk is converted into a numerical representation called an embedding. Embeddings capture the semantic meaning of the text, allowing for comparisons of similarity between different chunks.
4. **Saving:** Finally, the processed data, including the original text chunks, their source URLs, and the newly created embeddings, is saved to a JSON file in your Google Drive for later use.
In essence, the notebook takes raw website text and transforms it into a structured, semantically rich dataset.
## Setup
1. Open the provided Google Colab notebook.
2. Install the necessary libraries by running the first code cell:
bash !pip install -q ipywidgets google-colab python-docx pypdf pandas nltk sentence-transformers torch tqdm pyarrow httpx beautifulsoup4 datasets requests
3. Mount your Google Drive to save the output:
python from google.colab import drive drive.mount('/content/drive')
## Usage
1. **Web Scraping:** Run the second code cell to start the web scraping process. The script will crawl the specified `start_url` and collect text data from the linked pages within the same domain. You can adjust `max_pages` in the `crawl_website` function to control the number of pages scraped.
2. **Data Processing and Embedding:** Run the third code cell to process the scraped data into chunks and generate embeddings using the 'all-MiniLM-L6-v2' Sentence Transformer model.
3. **Save Embedded Data:** Run the fourth code cell to save the processed and embedded data to a JSON file named `embedded_dataset.json` in the `Output` folder of your Google Drive.
## Output
The notebook will generate an `embedded_dataset.json` file in your Google Drive containing a list of dictionaries. Each dictionary represents a text chunk and includes:
* `text`: The text content of the chunk.
* `source`: The URL from which the chunk was extracted.
* `embedding`: A list of floating-point numbers representing the embedding of the text chunk.
This dataset can be used for various downstream tasks such as building a question-answering system or implementing semantic search.
## Customization
* **Start URL:** Modify the `start_url` variable in the web scraping script to crawl a different website.
* **Chunking Parameters:** Adjust `chunk_size` and `chunk_overlap` in the `chunk_text` function to control how the text is split.
* **Sentence Transformer Model:** Change the model name in the `SentenceTransformer` constructor to use a different pre-trained model.
* **Output Path:** Modify the `

README.md

@@ -1,3 +1,31 @@
----
-license: mit
----
+---
+license: mit
+task_categories:
+- text-classification
+- feature-extraction
+- fill-mask
+- sentence-similarity
+- summarization
+- depth-estimation
+- object-detection
+- reinforcement-learning
+- robotics
+- tabular-classification
+- tabular-regression
+- tabular-to-text
+- table-to-text
+- multiple-choice
+- text-ranking
+- text-retrieval
+- time-series-forecasting
+language:
+- en
+tags:
+- code
+- chemistry
+- legal
+- art
+pretty_name: Microsoft Learn Dataset
+size_categories:
+- n>1T
+---