app.py

import streamlit as st
import pandas as pd
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.decomposition import LatentDirichletAllocation
import matplotlib.pyplot as plt

# Title and tabs
st.title("Topic Modeling for News Articles")
tab1, tab2 = st.tabs(["LDA Topic Modeling", "About the Dataset"])

# First Tab: Topic Modeling
with tab1:
    st.header("Input Data")
    
    # Text input for articles
    st.write("Paste your news articles (one article per line):")
    user_input = st.text_area("Enter articles here", height=200)
    
    if st.button("Analyze Topics"):
        if user_input.strip():
            # Convert input into a list of articles
            articles = user_input.split("\n")
            articles = [article.strip() for article in articles if article.strip()]
            
            # TF-IDF Vectorization
            vectorizer = TfidfVectorizer(stop_words='english', max_features=5000)
            tfidf_matrix = vectorizer.fit_transform(articles)
            
            # LDA Topic Modeling
            lda = LatentDirichletAllocation(n_components=5, random_state=42)
            lda.fit(tfidf_matrix)
            
            # Display topics
            st.subheader("Identified Topics")
            feature_names = vectorizer.get_feature_names_out()
            for idx, topic in enumerate(lda.components_):
                st.write(f"**Topic {idx + 1}:**", ", ".join([feature_names[i] for i in topic.argsort()[-10:]]))
            
            # Visualize topic distribution
            st.subheader("Topic Distribution")
            topic_distribution = lda.transform(tfidf_matrix)
            plt.figure(figsize=(10, 5))
            plt.bar(range(len(topic_distribution[0])), topic_distribution[0])
            plt.xlabel("Topics")
            plt.ylabel("Contribution")
            plt.title("Topic Distribution for the First Article")
            st.pyplot(plt.gcf())
        else:
            st.warning("Please input some articles to analyze.")

# Second Tab: About the Dataset
with tab2:
    st.header("About")
    st.write("This app performs topic modeling on news articles using Latent Dirichlet Allocation (LDA).")
    st.write("Paste articles in the text area, and the app will identify underlying topics.")