changes in the topic extraction model

app.py

@@ -26,24 +26,23 @@ collection = get_mongo_client()
 #     all_data = list(collection.find({}, {"_id": 0}))
 #     st.write(pd.DataFrame(all_data))
 
-#### **4. AI Chatbot with Sentiment Analysis & Topic Extraction**
-st.subheader("🤖 AI Chatbot with Sentiment Analysis & Topic Extraction")
+#### **4. AI Chatbot with Sentiment & Topic Analysis**
+st.subheader("🤖 AI Chatbot with Sentiment & Topic Analysis")
 
 # User input for chatbot
-user_prompt = st.text_area("Ask AI something or paste text for sentiment analysis:")
+user_prompt = st.text_area("Ask AI something or paste text for analysis:")
 
 if st.button("Analyze Sentiment & Get AI Response"):
-    ai_response, sentiment_label, confidence, topics = chatbot_response(user_prompt)
+    ai_response, sentiment_label, sentiment_confidence, topic_label, topic_confidence = chatbot_response(user_prompt)
 
     if ai_response:
         st.write("### AI Response:")
         st.write(ai_response)
 
         st.write("### Sentiment Analysis:")
-        st.write(f"**Sentiment:** {sentiment_label} ({confidence:.2f} confidence)")
-
-        st.write("### Extracted Topics:")
-        st.write(", ".join(topics) if topics else "No topics identified.")
+        st.write(f"**Sentiment:** {sentiment_label} ({sentiment_confidence:.2f} confidence)")
 
+        st.write("### Topic Extraction:")
+        st.write(f"**Detected Topic:** {topic_label} ({topic_confidence:.2f} confidence)")
     else:
         st.warning("⚠️ Please enter a question or text for analysis.")

chatbot.py

@@ -2,7 +2,6 @@ import os
 import streamlit as st
 import google.generativeai as genai
 from transformers import pipeline, AutoModelForSequenceClassification, AutoTokenizer
-from keybert import KeyBERT  # Topic Extraction
 
 # 🔑 Fetch API key from Hugging Face Secrets
 GEMINI_API_KEY = os.getenv("gemini_api")
@@ -12,7 +11,7 @@ if GEMINI_API_KEY:
 else:
     st.error("⚠️ Google API key is missing! Set it in Hugging Face Secrets.")
 
-# Correct Model Path
+# Model for Sentiment Analysis
 MODEL_NAME = "cardiffnlp/twitter-roberta-base-sentiment"
 
 # Load Sentiment Analysis Model
@@ -22,8 +21,17 @@ try:
 except Exception as e:
     st.error(f"❌ Error loading sentiment model: {e}")
 
-# Load KeyBERT for topic extraction
-kw_model = KeyBERT()
+# Load Topic Extraction Model
+try:
+    topic_pipeline = pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
+except Exception as e:
+    st.error(f"❌ Error loading topic extraction model: {e}")
+
+# Predefined topic labels for classification
+TOPIC_LABELS = [
+    "Technology", "Politics", "Business", "Sports", "Entertainment",
+    "Health", "Science", "Education", "Finance", "Travel", "Food"
+]
 
 # Function to analyze sentiment
 def analyze_sentiment(text):
@@ -42,18 +50,20 @@ def analyze_sentiment(text):
     except Exception as e:
         return f"Error analyzing sentiment: {e}", None
 
-# Function to extract key topics
-def extract_topics(text, num_keywords=3):
+# Function to extract topic
+def extract_topic(text):
     try:
-        keywords = kw_model.extract_keywords(text, keyphrase_ngram_range=(1, 2), top_n=num_keywords)
-        return [word[0] for word in keywords]  # Return only the keywords
+        topic_result = topic_pipeline(text, TOPIC_LABELS)
+        top_topic = topic_result["labels"][0]  # Get the highest confidence topic
+        confidence = topic_result["scores"][0]  # Confidence score for the topic
+        return top_topic, confidence
     except Exception as e:
-        return [f"Error extracting topics: {e}"]
+        return f"Error extracting topic: {e}", None
 
-# Function to generate AI response, analyze sentiment, and extract topics
+# Function to generate AI response, sentiment, and topic
 def chatbot_response(user_prompt):
     if not user_prompt:
-        return None, None, None, None
+        return None, None, None, None, None
 
     try:
         # AI Response from Gemini
@@ -61,11 +71,11 @@ def chatbot_response(user_prompt):
         ai_response = model.generate_content(user_prompt)
 
         # Sentiment Analysis
-        sentiment_label, confidence = analyze_sentiment(user_prompt)
+        sentiment_label, sentiment_confidence = analyze_sentiment(user_prompt)
 
         # Topic Extraction
-        topics = extract_topics(user_prompt)
+        topic_label, topic_confidence = extract_topic(user_prompt)
 
-        return ai_response.text, sentiment_label, confidence, topics
+        return ai_response.text, sentiment_label, sentiment_confidence, topic_label, topic_confidence
     except Exception as e:
-        return f"❌ Error: {e}", None, None, None
+        return f"❌ Error: {e}", None, None, None, None