Biomedical NER Application

A simple web application for biomedical Named Entity Recognition (NER) with user feedback capabilities.

Features

• Input biomedical text and get entity predictions
• View highlighted entities with their labels
• Rate prediction quality (1-5 stars)
• Add optional remarks to ratings
• View average rating across all predictions

Setup

1. Create a virtual environment (recommended):

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

2. Install dependencies:

pip install -r requirements.txt

3. Run the application:

python main.py

The application will be available at http://localhost:8000

API Endpoints

• GET /: Serves the web interface
• POST /predict: Submit text for NER analysis
• POST /rate: Submit feedback for a prediction
• GET /average-rating: Get the average rating across all predictions

Development

The application uses:

• FastAPI for the backend
• Tailwind CSS for styling
• HTMX for dynamic interactions
• JSON file storage for simplicity

Future Enhancements

• User authentication
• Database integration
• Advanced analytics
• Real NER model integration