Hugging Face's logo

Spaces:
NiranjanSathish
/
DrugBot_Retrieval_Based
Runtime error

App Files Community
DrugBot_Retrieval_Based / README.md
NiranjanSathish's picture
NiranjanSathish
Update README.md
541c7f3 verified
preview code
|
raw
history blame contribute delete
1.48 kB

A newer version of the Gradio SDK is available: 6.6.0

Upgrade
metadata 
title: Medical Drug QA Chatbot
emoji: 💊
colorFrom: blue
colorTo: indigo
sdk: gradio
sdk_version: 4.44.0
app_file: Scripts/app.py
pinned: false

💊 Medical Drug QA Chatbot

An intelligent chatbot that answers questions about medications using advanced NLP techniques.

Features

  • 🔍 Smart Query Processing: BioBERT-based NER for drug entity extraction
  • 📚 Hybrid Retrieval: FAISS + BioBERT semantic reranking
  • 🤖 AI-Powered Answers: Groq Llama-4 for natural language generation
  • 💾 Comprehensive Database: Mayo Clinic drug information

Usage

Simply ask questions about:

  • Side effects and warnings
  • Dosage and usage instructions
  • Drug interactions
  • Storage guidelines
  • Precautions for specific conditions

Example Questions

  • "What are the side effects of Aspirin?"
  • "How should I store Insulin?"
  • "What precautions should I take with Lisinopril?"
  • "Can I take Metformin with alcohol?"

Tech Stack

  • Frontend: Gradio
  • NER: BioBERT (alvaroalon2/biobert_chemical_ner)
  • Embeddings: MiniLM-V6, BioBERT
  • Vector DB: FAISS
  • LLM: Llama-4 via Groq API

⚠️ Disclaimer

This chatbot provides educational information only. Always consult healthcare professionals for medical advice.

Setup

  1. Clone the repository
  2. Install dependencies: pip install -r requirements.txt
  3. Set GROQ_API_KEY environment variable
  4. Build FAISS index: python Scripts/Retrieval.py
  5. Run: python app.py