AI-Powered Document Search and Analysis

This is an AI-powered web application that allows users to upload documents, perform semantic searches using large language models (LLMs), and generate document summaries.

Features

• Document Upload: Upload documents for search and analysis.
• Semantic Search: Perform AI-based contextual searches using LLMs.
• Chat with Documents: Chat with your documents using LLMs.
• Text Summarization: Generate concise summaries of document content.
• User Authentication: Secure login and registration with JWT tokens.

Tech Stack

• Backend: Python (FastAPI)
• Frontend: Svelte
• Database: PostgreSQL
• AI Libraries: LangChain, OpenAI
• Containerization: Docker, Docker Compose

Installation

1. Clone the repository:

   git clone https://github.com/Leg0shii/smart-documents.git
   cd smart-documents
   

2. Set up the environment variables in backend/.env:

   FRONTEND_PORT=5000
   BACKEND_PORT=8000
   SECRET_KEY=your_secret_key
   OPENAI_API_KEY=your_openai_api_key
   POSTGRES_USER=postgres
   POSTGRES_PASSWORD=postgres
   POSTGRES_DB=smart_documents
   POSTGRES_HOST=db
   POSTGRES_PORT=5432
   DATABASE_URL=postgresql://${POSTGRES_USER}:${POSTGRES_PASSWORD}@${POSTGRES_HOST}:${POSTGRES_PORT}/${POSTGRES_DB}
   

3. Run the application with Docker:

   docker-compose up --build
   

4. Access the application:

   • Frontend: http://localhost:5000
   • Backend: http://localhost:8000

Usage

• Register/Login: Create an account or log in.
• Upload Documents: Upload your documents for search.
• Perform Search: Use the search bar to find documents with custom top K results.
• Get Summaries: Retrieve summaries of the search results.

Development

To run locally without Docker:

• Backend:

  cd backend
  pip install -r requirements.txt
  uvicorn app.main:app --reload
  

• Frontend:

  cd frontend
  npm install
  npm run dev
  

License

This project is licensed under the MIT License.