<<<<<<< HEAD

LLM_RAG_CHATBOT

=======

Document Chat Application

A complete end-to-end web application that allows users to chat with an AI assistant restricted only to the documents they upload. Built with Angular frontend, Spring Boot backend, and Gemini Pro API.

Features

• Document Upload: Support for PDF, DOCX, and TXT files
• Text Extraction: Automatic text extraction from uploaded documents
• Vector Search: Document chunks are embedded and stored for semantic search
• AI Chat: Chat interface that answers only from uploaded documents
• Smart Responses:
  • Answers from documents when relevant
  • "Out of scope" for unrelated questions
  • "No documents available" when no documents are uploaded
  • Highlights conflicts when documents contain conflicting information

Tech Stack

• Frontend: Angular 17
• Backend: Spring Boot 3.2.0
• Database: H2 (in-memory)
• AI: Google Gemini Pro API
• Document Processing: Apache PDFBox, Apache POI
• Vector Storage: Custom implementation with cosine similarity

Quick Start

Prerequisites

• Java 17 or higher
• Node.js 18 or higher
• npm or yarn
• Google Gemini API key (optional - app works without it using dummy responses)

1. Backend Setup

cd backend
./mvnw spring-boot:run

The backend will start on http://localhost:8080

2. Frontend Setup

cd frontend
npm install
npm start

The frontend will start on http://localhost:4200

3. Configure Gemini API (Required for AI Responses)

⚠️ IMPORTANT: The application requires a Gemini API key to work properly. Without it, you'll get dummy responses.

Quick Setup:

./setup-api-key.sh

Manual Setup:

1. Get your API key:

   • Go to: https://makersuite.google.com/app/apikey
   • Create a new API key
   • Copy the API key

2. Set the API key (choose one method):

   Option A - Environment Variable (Recommended):

   export GEMINI_API_KEY=your_api_key_here

   Option B - Add to application.properties:

   gemini.api.key=your_api_key_here

   Option C - Set temporarily for testing:

   GEMINI_API_KEY=your_api_key_here ./start-backend.sh

3. Restart the backend after setting the API key:

   cd backend && mvn spring-boot:run

How It Works

1. Document Upload: Users upload PDF, DOCX, or TXT files through the web interface
2. Text Extraction: Backend extracts text from documents using Apache PDFBox and POI
3. Chunking: Text is split into manageable chunks (1000 characters with 200 character overlap)
4. Embedding: Each chunk is converted to a vector embedding using Gemini's embedding API
5. Storage: Embeddings are stored in the database with the original text
6. Query Processing: When users ask questions:
   • Query is converted to an embedding
   • Similar chunks are found using cosine similarity
   • Relevant context is sent to Gemini Pro for response generation
   • Response is returned with source information

API Endpoints

Documents

• POST /api/documents/upload - Upload a document
• GET /api/documents - Get all documents
• DELETE /api/documents/{id} - Delete a document

Chat

• POST /api/chat/message - Send a chat message

Configuration

Backend Configuration (application.properties)

# Server
server.port=8080

# Database
spring.datasource.url=jdbc:h2:mem:testdb
spring.datasource.username=sa
spring.datasource.password=password

# File upload
spring.servlet.multipart.max-file-size=10MB
spring.servlet.multipart.max-request-size=10MB

# Gemini API
gemini.api.key=${GEMINI_API_KEY:}

Development

Backend Development

cd backend
./mvnw spring-boot:run

Frontend Development

cd frontend
npm start

Building for Production

Backend:

cd backend
./mvnw clean package
java -jar target/document-chat-backend-0.0.1-SNAPSHOT.jar

Frontend:

cd frontend
npm run build

Troubleshooting

Common Issues

1. Port already in use: Change the port in application.properties
2. CORS errors: Ensure the frontend URL is correct in the CORS configuration
3. File upload fails: Check file size limits and supported formats
4. No AI responses or "Out of scope" for valid questions:
   • Most common cause: Missing or invalid Gemini API key
   • Run ./setup-api-key.sh to check your API key setup
   • Verify the API key is set: echo $GEMINI_API_KEY
   • Restart the backend after setting the API key
5. Poor response quality: The dummy embedding system has limitations; use a real Gemini API key for better results

Logs

Backend logs are available in the console. For more detailed logging, modify logback-spring.xml.

License

This project is for educational purposes. Please ensure you comply with Google's Gemini API terms of service when using the AI features.

0822860 (RAG Chatbot)