AWS Builder Center
From Words to Wisdom: Revolutionizing Graph Data Modelling with GenAI and Natural Language

From Words to Wisdom: Revolutionizing Graph Data Modelling with GenAI and Natural Language

Join Neptune specialists as they unveil a ground-breaking solution that bridges the gap between human communication and graph data modelling. In this exciting session, you'll discover how the power of Generative AI and natural language processing is transforming the traditionally complex task of graph database design into an intuitive, conversation-driven experience.

This episode of "Let's Talk About Data" featured Brian O'Keefe and Kevin Phillips discussing how GenAI can revolutionize graph database modeling for Amazon Neptune. The conversation centered on a new toolkit that uses natural language to dramatically simplify the traditionally complex process of creating graph models, reducing what typically takes days or weeks down to just minutes.
The speakers explained that graph databases excel at handling connected data and are ideal for use cases like fraud detection, knowledge management, cybersecurity, supply chain logistics, and social networks. Amazon Neptune offers two services: Neptune Database for transactional operations and Neptune Analytics for in-memory analytics workloads. The main challenge customers face is the mindset shift from relational database thinking to relationship-focused modeling, which requires understanding how to structure data as nodes and edges.
The GenAI solution demonstrated during the show allows users to describe their use case in plain language, provide three key questions they want answered, and automatically generates graph schemas, sample data, and queries. Built on AWS services including CloudFront, Cognito, API Gateway, Lambda, DynamoDB, and Bedrock, the tool won an internal AWS hackathon and is currently in security review before being released as open source on GitHub. Future enhancements include agentic AI workflows, improved data generation, natural language query conversion, and automatic query optimization.
Key Highlights:
  • Graph databases focus on relationships between data points, making them ideal for fraud detection, knowledge management, and social network analysis
  • Amazon Neptune provides both transactional (Neptune Database) and analytics (Neptune Analytics) capabilities with support for OpenCypher, Gremlin, and SPARQL query languages
  • GenAI modeling toolkit reduces graph modeling time from weeks to minutes through natural language interaction
  • Core features include automated schema generation, sample data creation, visual model representation, and export to Neptune-compatible formats
  • Architecture leverages React frontend, Amazon Cognito authentication, Lambda backend, DynamoDB storage, and Amazon Bedrock for AI capabilities
  • Future roadmap includes agentic AI workflows, enhanced data generation, natural language to query conversion, and performance optimization
  • Open source release planned after security review, allowing customers to deploy in their own AWS accounts
  • Democratization goal aims to make graph database technology accessible beyond expert users


Hosts of the show 🎤

 Ibrahim Emara  - Database Specialist SA @AWS

Guests 🎤

 Kevin Phillips  - Amazon Neptune SR Specialist SA @AWS
 Brian O'Keefe  - Amazon Neptune Principal Specialist SA @AWS
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article