Spaces:
Sleeping
Sleeping
File size: 2,824 Bytes
d19998a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 |
import streamlit as st
from transformers import pipeline
from wordcloud import WordCloud
import matplotlib.pyplot as plt
from huggingface_hub import login
hf_token = st.secrets.get("HF_TOKEN")
if hf_token:
login(hf_token)
st.set_page_config(page_title="📰 News Veracity Checker", layout="wide")
@st.cache_resource
def load_models():
veracity_model = pipeline('text-classification', model='jy46604790/Fake-News-Bert-Detect')
summarizer = pipeline('summarization', model='facebook/bart-large-cnn')
explainer = pipeline('text-generation', model='google/gemma-2b-it', max_new_tokens=200)
return veracity_model, summarizer, explainer
veracity_model, summarizer, explainer = load_models()
def get_summary(text):
summary = summarizer(text, max_length=120, min_length=30, do_sample=False)
return summary[0]['summary_text']
def generate_explanation(text, veracity_label):
prompt = (f"You are a news analyst. Given this article:\n\n{text}\n\n"
f"Explain clearly why this news is {veracity_label.lower()}. "
f"If false, provide an accurate version or correct factual information. Explanation:")
response = explainer(prompt)[0]['generated_text']
return response.split("Explanation:")[-1].strip()
def plot_wordcloud(text):
wc = WordCloud(width=800, height=400, background_color='white').generate(text)
plt.figure(figsize=(10, 5))
plt.imshow(wc, interpolation='bilinear')
plt.axis('off')
st.pyplot(plt)
def get_veracity(text):
res = veracity_model(text[:512])[0]
label = res['label']
score = round(res['score'], 3)
if label == "LABEL_0":
return "False", score
else:
return "True", score
#Streamlit
st.title("📰 News Veracity Checker")
st.sidebar.info("Paste a news headline or article to get a veracity check, summary, and AI-generated explanation.")
input_text = st.text_area("Paste news headline or article:", height=200)
if st.button("🚀 Analyze"):
with st.spinner("Analyzing the content..."):
label, score = get_veracity(input_text)
if label == "True":
st.markdown(f"""<h4>✅ <span style='color:green;'>Veracity: True</span> ({score*100:.2f}% confidence)</h4>""",unsafe_allow_html=True)
else:
st.markdown(f"""<h4>❌ <span style='color:red;'>Veracity: False</span> ({score*100:.2f}% confidence)</h4>""",unsafe_allow_html=True)
#summary
summary = get_summary(input_text)
st.markdown("### 📃 **Summary:**")
st.info(summary)
#explanation
explanation = generate_explanation(input_text, label)
st.markdown("### 🤖 **AI Explanation:**")
st.write(explanation)
plot_wordcloud(input_text)
|