try4

app.py

@@ -144,6 +144,9 @@ def main():
                 gemini_result = analyze_content_gemini(gemini_model, news_text)
 
                 # First show the high-level metrics in columns
+                # In the main() function:
+
+                # Display metrics in columns
                 col1, col2, col3 = st.columns(3)
 
                 with col1:
@@ -163,83 +166,51 @@ def main():
                     st.metric("Prediction", gemini_pred)
                     st.metric("Confidence", f"{gemini_conf}%")
 
-                # Single expander for all detailed analysis
-                with st.expander("View All Analysis Details"):
-                    # Named Entities Section
-                    st.subheader("🏷️ Named Entities")
-                    entities = extract_entities(news_text, nlp)
-                    df = pd.DataFrame(entities, columns=["Entity", "Type"])
-                    st.dataframe(df)
-                    
-                    # Knowledge Graph Section
-                    st.subheader("🕸️ Knowledge Graph")
-                    fig = generate_knowledge_graph_viz(news_text, nlp, tokenizer, model)
-                    st.plotly_chart(fig, use_container_width=True)
-                    
-                    # Detailed Gemini Analysis
-                    st.subheader("📝 Text Classification")
-                    text_class = gemini_result.get('text_classification', {})
-                    st.write(f"Category: {text_class.get('category', 'N/A')}")
-                    
-                # Detailed analysis sections
-                with st.expander("View Detailed Analysis"):
-                    with st.expander("View Detailed Analysis"):
-                        try:
-                            # Text Classification
-                            st.subheader("📝 Text Classification")
-                            text_class = gemini_result.get('text_classification', {})
-                            st.write(f"Category: {text_class.get('category', 'N/A')}")
-                            st.write(f"Writing Style: {text_class.get('writing_style', 'N/A')}")
-                            st.write(f"Target Audience: {text_class.get('target_audience', 'N/A')}")
-                            st.write(f"Content Type: {text_class.get('content_type', 'N/A')}")
-                            
-                            # Sentiment Analysis
-                            st.subheader("🎭 Sentiment Analysis")
-                            sentiment = gemini_result.get('sentiment_analysis', {})
-                            st.write(f"Primary Emotion: {sentiment.get('primary_emotion', 'N/A')}")
-                            st.write(f"Emotional Intensity: {sentiment.get('emotional_intensity', 'N/A')}/10")
-                            st.write(f"Sensationalism Level: {sentiment.get('sensationalism_level', 'N/A')}")
-                            st.write("Bias Indicators:", ", ".join(sentiment.get('bias_indicators', ['N/A'])))
+                # Single expander for all analysis details
+                with st.expander("Detailed Analysis"):
+                    try:
+                        # Text Classification
+                        st.subheader("📝 Text Classification")
+                        text_class = gemini_result.get('text_classification', {})
+                        st.write(f"Category: {text_class.get('category', 'N/A')}")
+                        st.write(f"Writing Style: {text_class.get('writing_style', 'N/A')}")
+                        st.write(f"Target Audience: {text_class.get('target_audience', 'N/A')}")
+                        st.write(f"Content Type: {text_class.get('content_type', 'N/A')}")
+                        
+                        # Sentiment Analysis
+                        st.subheader("🎭 Sentiment Analysis")
+                        sentiment = gemini_result.get('sentiment_analysis', {})
+                        st.write(f"Primary Emotion: {sentiment.get('primary_emotion', 'N/A')}")
+                        st.write(f"Emotional Intensity: {sentiment.get('emotional_intensity', 'N/A')}/10")
+                        st.write(f"Sensationalism Level: {sentiment.get('sensationalism_level', 'N/A')}")
+                        st.write("Bias Indicators:", ", ".join(sentiment.get('bias_indicators', ['N/A'])))
+                        
+                        # Entity Recognition
+                        st.subheader("🔍 Entity Recognition")
+                        entities = gemini_result.get('entity_recognition', {})
+                        st.write(f"Source Credibility: {entities.get('source_credibility', 'N/A')}")
+                        st.write("People:", ", ".join(entities.get('people', ['N/A'])))
+                        st.write("Organizations:", ", ".join(entities.get('organizations', ['N/A'])))
+                        
+                        # Named Entities from spaCy
+                        st.subheader("🏷️ Named Entities")
+                        entities = extract_entities(news_text, nlp)
+                        df = pd.DataFrame(entities, columns=["Entity", "Type"])
+                        st.dataframe(df)
+                        
+                        # Knowledge Graph Visualization
+                        st.subheader("🕸️ Knowledge Graph")
+                        fig = generate_knowledge_graph_viz(news_text, nlp, tokenizer, model)
+                        st.plotly_chart(fig, use_container_width=True)
+                        
+                        # Analysis Reasoning
+                        st.subheader("💭 Analysis Reasoning")
+                        for point in gemini_result.get('gemini_analysis', {}).get('reasoning', ['N/A']):
+                            st.write(f"• {point}")
                             
-                            tone = sentiment.get('tone', {})
-                            st.write(f"Tone Formality: {tone.get('formality', 'N/A')}")
-                            st.write(f"Tone Style: {tone.get('style', 'N/A')}")
-                            st.write("Emotional Triggers:", ", ".join(sentiment.get('emotional_triggers', ['N/A'])))
-                            
-                            # Entity Recognition
-                            st.subheader("🔍 Entity Recognition")
-                            entities = gemini_result.get('entity_recognition', {})
-                            st.write(f"Source Credibility: {entities.get('source_credibility', 'N/A')}")
-                            st.write("People:", ", ".join(entities.get('people', ['N/A'])))
-                            st.write("Organizations:", ", ".join(entities.get('organizations', ['N/A'])))
-                            st.write("Locations:", ", ".join(entities.get('locations', ['N/A'])))
-                            st.write("Dates:", ", ".join(entities.get('dates', ['N/A'])))
-                            st.write("Statistics:", ", ".join(entities.get('statistics', ['N/A'])))
-                            
-                            # Context
-                            st.subheader("📰 Context")
-                            context = gemini_result.get('context', {})
-                            st.write("Main Narrative:", context.get('main_narrative', 'N/A'))
-                            st.write("Supporting Elements:", ", ".join(context.get('supporting_elements', ['N/A'])))
-                            st.write("Key Claims:", ", ".join(context.get('key_claims', ['N/A'])))
-                            st.write("Narrative Structure:", context.get('narrative_structure', 'N/A'))
-                            
-                            # Fact Checking
-                            st.subheader("✔️ Fact Checking")
-                            fact_check = gemini_result.get('fact_checking', {})
-                            st.write("Verifiable Claims:")
-                            for claim in fact_check.get('verifiable_claims', ['N/A']):
-                                st.write(f"• {claim}")
-                            st.write(f"Evidence Present: {fact_check.get('evidence_present', 'N/A')}")
-                            st.write(f"Fact Check Score: {fact_check.get('fact_check_score', 'N/A')}/100")
-                            
-                            # Analysis Reasoning
-                            st.subheader("💭 Analysis Reasoning")
-                            for point in gemini_result.get('gemini_analysis', {}).get('reasoning', ['N/A']):
-                                st.write(f"• {point}")
-                                
-                        except Exception as e:
-                            st.error("Error processing Gemini analysis results")
+                    except Exception as e:
+                        st.error("Error processing analysis results")
+
 
 
                 with st.expander("Named Entities"):