National Industrial Classification (NIC) Code Semantic Search

This project provides a semantic search system for National Industrial Classification (NIC) codes, leveraging ModernBERT embeddings and Faiss for efficient similarity search. The backend is powered by FastAPI, offering a robust and fast API, while the frontend provides a user-friendly interface with both text and voice input capabilities.

Features

• ModernBERT-based Semantic Search: Utilizes nomic-ai/modernbert-embed-base for generating high-quality embeddings, enabling accurate semantic search.
• Efficient Faiss Indexing: Employs Faiss for rapid similarity search, ensuring quick retrieval of relevant NIC codes.
• FastAPI Backend: A lightweight and high-performance API built with FastAPI, designed for efficient handling of search queries.
• User-Friendly Frontend: A simple and intuitive webpage with support for both text and voice input, enhancing user experience.

Prerequisites

Before setting up the project, ensure you have the following installed:

• Python 3.8+
• pip (Python package manager)
• Git (for cloning the repository)
• Node.js (optional, for frontend modifications)

Setup Instructions

1. Clone the Repository:

   git clone [https://github.com/YOUR_GITHUB_USERNAME/YOUR_REPOSITORY_NAME.git](https://github.com/YOUR_GITHUB_USERNAME/YOUR_REPOSITORY_NAME.git)
   cd YOUR_REPOSITORY_NAME

2. Install Backend Dependencies:

   • Create a virtual environment (recommended):

     python -m venv venv
     source venv/bin/activate  # On macOS/Linux
     venv\Scripts\activate  # On Windows

   • Install the required Python packages:

     pip install -r requirements.txt

3. Place Data Files:

   • Ensure that modernbert_faiss_index.bin and combined.csv are located in the same directory as app.py.

Running the Backend (FastAPI)

1. Start the FastAPI server:

   uvicorn app:app --host 0.0.0.0 --port 9040

   • The API will be accessible at: http://127.0.0.1:9040
   • Swagger UI documentation: http://127.0.0.1:9040/docs

Running the Frontend

1. Open index.html in your web browser.

   • The frontend will interact with the backend API to perform search queries.

API Endpoints

1. Search Endpoint

• URL: /search

• Method: POST

• Request Format (JSON):

  {
    "query": "Your search term",
    "k": 5
  }

• Response Example:

  {
    "query": "Your search term",
    "translated_query": "Translated text",
    "results": [
      {
        "S.No.": 1,
        "Description": "Industry description",
        "Class": "123",
        "Group": "456",
        "Sub Class": "789"
      },
      // ... more results
    ]
  }

2. Root Endpoint

• URL: /

• Method: GET

• Response:

  {
    "message": "Welcome to the Semantic Search API!"
  }

Troubleshooting

• Faiss Index Not Found: Verify that modernbert_faiss_index.bin exists in the correct directory.
• CORS Issues: Ensure that the API allows requests from your frontend's origin.
• Missing Dependencies: Re-run pip install -r requirements.txt.
• Port Conflicts: Change the port in uvicorn app:app --port 9040 to an available port.

Contributing

Contributions are welcome! Feel free to fork the repository, submit issues, or open pull requests.

License

This project is licensed under the MIT License.