app.py

import streamlit as st
import pandas as pd
from db import insert_data_if_empty, get_mongo_client
from chatbot import chatbot_response

# Sidebar: Display image and title.
st.sidebar.image("https://huggingface.co/spaces/sharangrav24/SentimentAnalysis/resolve/main/sen_analysis.png", width=200)
st.sidebar.markdown("## Group A Submission - Python")

# Sidebar: Add submitted by details.
st.sidebar.markdown("""
**Submitted by-**  
📌Kumar Sharangrav [C] (GMP-21-10)  
📌Amit Sanjeev (GMP-21-01)  
📌Anoop G Zacharia (GMP-21-03)  
📌Anviti Pant (GMP-21-05)
""")

# Ensure historical data is inserted into MongoDB if not already present.
insert_data_if_empty()

# Connect to MongoDB (optional: for additional visualizations)
collection = get_mongo_client()

st.subheader("💬 Chatbot with Sentiment Analysis & Category Extraction")

# Create an expander to display example questions on separate lines.
with st.expander("👋 Hi, allow me to help you with prompts:"):
    st.write("💡 Provide analysis for data entry 1 in the dataset")
    st.write("💡 What is the dataset summary?")
    st.write("💡 or just ask me something of your own, I'll be happy to help 😊")

# Text area for user input.
user_prompt = st.text_area("Ask me something:")

if st.button("Get Response"):
    ai_response, sentiment_label, sentiment_confidence, topic_label, topic_confidence = chatbot_response(user_prompt)
    if ai_response:
        st.write("### Response:")
        st.markdown(ai_response)
        st.write("### Sentiment Analysis:")
        st.write(f"**Sentiment Detected:** {sentiment_label} ({sentiment_confidence:.2f} confidence)")
        st.write("### Category Extraction:")
        st.write(f"**Category Detected:** {topic_label} ({topic_confidence:.2f} confidence)")
else:
        st.warning("⚠️ Please enter a question or text for analysis.")