This project explores the synergy between Neo4j and Large Language Models (LLMs), offering solutions for natural language interfaces to knowledge graphs, knowledge graph creation from unstructured data, and report generation. It targets developers and researchers interested in leveraging graph databases with advanced AI capabilities.

How It Works

The project demonstrates three core use cases: a natural language interface to a knowledge graph, knowledge graph construction from unstructured text, and report generation using both static data and LLM outputs. It utilizes Neo4j as the graph database and integrates LLM functionalities within its backend. The architecture separates backend and frontend code, with distinct React applications for each use case.

Quick Start & Requirements

• Install and run via docker-compose up after creating an .env file from env.example.
• Requires Docker.
• Demo UI: http://localhost:4173/
• Demo Backend: localhost:7860
• Demo Neo4j Database: neo4j+s://demo.neo4jlabs.com (username: companies, password: companies, database: companies)
• Blog posts detailing learnings: https://medium.com/neo4j/

Highlighted Details

• Focuses on three primary use cases: NL Interface to KG, KG Creation from Unstructured Data, and Report Generation.
• Backend code organized in api/, frontend in ui/src/ with separate React apps per use case.
• Demo database is a subset of the Diffbot knowledge graph, containing company, people, and article data with text embeddings.

Maintenance & Community

The project is from Neo4j, with blog posts indicating active exploration and learning in the field. Contributions are welcomed via issues and pull requests.

Licensing & Compatibility

The repository does not explicitly state a license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

The project acknowledges the rapidly evolving nature of AI and LLMs, stating that information, assumptions, and code are based on current understanding and subject to change.