Welcome to this week’s newsletter! You’ll find a couple of items using ChatGPT as a developer assistant – but from different starting points. In “Generating Cypher Queries With ChatGPT 4 on Any Graph Schema”, Tomaz Bratanic creates an experimental model in Python that generates Cypher from existing schemas. In a separate video tutorial, “ChatGPT Generated Star Wars Data,” Jonathan Thein uses ChatGPT to create a database of movie characters, then lets a human query the database.
If you are interested in using graph databases as part of AWS architectures, read “When to Use a Graph Database Like Neo4j on AWS”, covering patterns for streaming, batch, Graph Data Science-enhanced feature engineering, and knowledge graphs for domain-specific LLMs.
NODES 2023 Call for Papers is open until June 30. Don’t wait. You can be selected as a featured speaker on our website if you submit your talk by May 31.
FEATURED NODES SPEAKER: Roei Levi
In his NODES 2022 presentation, he introduces Cymple, a new open-source Python package. It creates neat, reusable Cypher queries with auto-completion in Python. Watch his talk!
NETWORKING: Twitch Graph Network Analysis Using Neo4jHarine explores a Twitch Gamers dataset, importing CSV data with Neo4j’s admin-import terminal command. Clusters are then formed on this dataset using GDS’s Louvain community detection function.
GENERATIVE AI: Generating Cypher Queries With ChatGPT 4 on Any Graph SchemaTomaz Bratanic continues to refine his techniques for translating natural language to Cypher. This blog post will show you how to implement a Cypher statement-generating model by providing only the graph schema information.
INTEGRATION: What’s New in ArcGIS Knowledge
TWEET OF THE WEEK: @rastadidiDon’t forget to retweet if you like it!
📢 @neo4j has a Ninja program… but who knows:— Adrien SALES (@rastadidi) May 1, 2023
🤔 The aim of this program
🥷 Who they are
🗺️ Where they come from ?
👭 Ninja women ?
💡 That's what you'll discover with this #datadriven dedicated blog post on @kaggle .
🎀 Live demo on @duckdb https://t.co/sHIHKG4aPR#jupyter