LLM-RAG-Chatbot

Description

This project implements a chatbot powered by Large Language Models (LLM) through Retrieval-Augmented Generation (RAG). RAG enhances LLM capabilities by retrieving relevant information from a database, enabling the generation of contextually relevant responses. Specifically, this chatbot is designed for use within a hospital system, capable of retrieving and providing information on patients, patient experiences, visits, insurance payers, and physicians.

Installation and Configuration

Follow these steps to set up the LLM-RAG-Chatbot:

1. Configure OpenAI API Key

Ensure you have an OpenAI account and a valid API key. Set up environment variables for OpenAI as follows:

2. Sign Up for Neo4j AuraDB

Create a free account on Neo4j AuraDB by clicking the Start Free button, followed by New Instance. After selecting Download and Continue, your instance will be created and should be up and running.

3. Create a .env File

In the project's root directory, create a .env file with the necessary configurations:

OPENAI_API_KEY="YOUR_OPENAI_API_KEY"
NEO4J_URI=YOUR_NEO4J_URI
NEO4J_USERNAME=YOUR_NEO4J_USERNAME
NEO4J_PASSWORD=YOUR_NEO4J_PASSWORD
HOSPITALS_CSV_PATH=https://raw.githubusercontent.com/JuansonGrajales/LLM-RAG-Chatbot/main/data/hospitals.csv
PAYERS_CSV_PATH=https://raw.githubusercontent.com/JuansonGrajales/LLM-RAG-Chatbot/main/data/payers.csv
PHYSICIANS_CSV_PATH=https://raw.githubusercontent.com/JuansonGrajales/LLM-RAG-Chatbot/main/data/physicians.csv
PATIENTS_CSV_PATH=https://raw.githubusercontent.com/JuansonGrajales/LLM-RAG-Chatbot/main/data/patients.csv
VISITS_CSV_PATH=https://raw.githubusercontent.com/JuansonGrajales/LLM-RAG-Chatbot/main/data/visits.csv
REVIEWS_CSV_PATH=https://raw.githubusercontent.com/JuansonGrajales/LLM-RAG-Chatbot/main/data/reviews.csv
HOSPITAL_AGENT_MODEL=gpt-3.5-turbo-1106
HOSPITAL_CYPHER_MODEL=gpt-3.5-turbo-1106
HOSPITAL_QA_MODEL=gpt-3.5-turbo-0125
CHATBOT_URL=http://host.docker.internal:8000/hospital-rag-agent

4. Start the Application

Open a terminal and execute the following command:

docker-compose up --build

After the build and run process completes, access the chatbot UI at http://localhost:8501/ to start interacting with the chatbot.

Source

The tutorial for building this chatbot is available at Real Python by Harrison Hoffmam