This project demonstrates how to perform graph-based modeling and information extraction from a Neo4j database using Large Language Models (LLMs) via LangChain. The focus is on querying health data stored in a Neo4j graph database and utilizing LLMs for question answering (QA).
- Neo4j Integration: Connects to a Neo4j database to access and query health-related graph data.
- LangChain Support: Leverages the LangChain library for seamless integration with LLMs.
- Question Answering: Provides answers to natural language queries on the health data graph using ChatOpenAI.
- Configurable Query Execution: Supports Cypher queries to extract relevant insights from the database.
- Python 3.8+
- Neo4j (Database must be set up and running)
- LangChain
- OpenAI API access
- Streamlit