Aegis 🤖: A Q&A Chatbot for Medical Information

This project is a medical Q&A chatbot named Aegis, built using LangChain, Hugging Face, and Groq. It's designed to answer questions based on a provided set of PDF documents. The chatbot uses a Retrieval-Augmented Generation (RAG) architecture to find relevant information in the documents before generating a response.

🌐 Live Demo

Try Aegis now: https://aegis--the-ai-medical-chatbot-l2mrvkzmomehccz2hvssv8.streamlit.app/

✨ Features

• Retrieval-Augmented Generation (RAG): Aegis retrieves information from a knowledge base (PDFs) to provide accurate answers.
• Fast and Efficient: Utilizes the high-performance Groq API for rapid inference.
• Persistent Memory: Uses FAISS for an efficient vector store to remember and retrieve information from the documents.
• Streamlit Interface: Provides a user-friendly web interface for interaction.

⚙️ How It Works

1. Load and Process Documents: PDF files from the data/ directory are loaded and split into smaller chunks.
2. Create Embeddings: The sentence-transformers/all-MiniLM-L6-v2 model is used to convert the text chunks into numerical vectors (embeddings).
3. Vector Store: These embeddings are stored in a FAISS vector database (vectorstore/db_faiss). This acts as the chatbot's long-term memory.
4. User Query: When a user asks a question, the query is also converted into an embedding.
5. Retrieval: Aegis retrieves the most semantically similar text chunks from the FAISS database.
6. Generation: These retrieved chunks, along with the user's question, are fed to a large language model (LLM) hosted by Groq to generate a final, contextually relevant answer.

🚀 Setup and Installation

Prerequisites

• Python 3.8+
• A Groq API Key and a Hugging Face API token.

Step 1: Clone the Repository

git clone https://github.com/your-username/your-repository-name.git
cd your-repository-name