Document Q&A System

A modern, AI-powered document question and answer system built with Python and OpenAI's API.

🚀 Features

• Modern GUI: Sleek CustomTkinter interface with dark theme
• Multiple AI Models: Choose from GPT-4o Mini, GPT-4o, GPT-3.5 Turbo, or GPT-4
• Smart PDF Processing: Intelligent text chunking with overlap for better context
• Vector Search: Advanced similarity search for relevant document sections
• Real-time Q&A: Interactive question answering with progress indicators

📁 Project Structure

document_searcher/
├── main.py                 # Original command-line interface
├── modern_gui.py          # Modern GUI application
├── run_gui.py             # GUI launcher script
├── query_manager.py       # Command-line query interface
├── qa_agent.py            # AI model interface
├── document_loader.py     # PDF processing and chunking
├── embeddings_manager.py  # Vector embeddings management
├── vector_search.py      # Similarity search
├── pdf_metadata.py       # PDF metadata extraction
├── model_comparison.py   # Model performance testing
├── gui_app.py            # Original GUI (legacy)
├── api_testing.py        # API testing utilities
├── view_usage.py         # Usage monitoring
└── README.md             # This file

🛠️ Installation

1. Clone the repository:

   git clone https://github.com/JamesKruik/document-qa-system.git
   cd document-qa-system/document_searcher

2. Install dependencies:

   pip install openai python-dotenv PyPDF2 numpy customtkinter

3. Set up environment variables:

   • Create a .env file in the document_searcher directory
   • Add your OpenAI API key:

     OPENAI_API_KEY=your_api_key_here
     

🎮 Usage

GUI Application (Recommended)

python modern_gui.py

Command Line Interface

python main.py

Query Manager (Interactive CLI)

python query_manager.py

Model Comparison

python model_comparison.py

🤖 Available AI Models

Model	Cost	Accuracy	Best For
GPT-4o Mini	Low	Good	Simple questions
GPT-4o	Medium	Excellent	Most tasks
GPT-3.5 Turbo	Very Low	Medium	Technical content
GPT-4	High	Outstanding	Complex reasoning

📖 How It Works

1. Load PDFs: Select and load your PDF documents
2. Process Documents: Create vector embeddings for intelligent search
3. Ask Questions: Type questions and get accurate answers
4. Choose Models: Select the AI model that fits your needs

🔧 Technical Details

• PDF Processing: PyPDF2 for text extraction
• Text Chunking: Smart chunking with sentence boundary detection
• Embeddings: OpenAI's text-embedding-3-small model
• Vector Search: Cosine similarity with multiple chunk retrieval
• GUI Framework: CustomTkinter for modern interface

📝 Example Questions

• "How do I configure the device settings?"
• "What are the technical specifications?"
• "What troubleshooting steps are available?"

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

📄 License

This project is open source and available under the MIT License.

🙏 Acknowledgments

• OpenAI for the powerful API
• CustomTkinter for the modern GUI framework
• PyPDF2 for PDF processing capabilities