Graph Data Science 2.3 was just released! Check out the new features like the Minimum Directed Steiner Tree, super useful for understanding the shortest or least expensive routes when travelling from multiple locations to a specific destination. Or HashGNN, where, instead of doing neural transformations like most GNNs, transformations are done by locality-sensitive min-hashing. We are also introducing negative relationships, which provide negative relationship examples for more options to train link prediction models and get them into production quickly. Learn more about it here.
P.S.: If you’re a developer building modern applications with GraphQL, don’t forget to take this short, two-minute survey. We want to hear from you!
Mark is an Apache Pinot Advocate and Developer Relations Engineer at StarTree. He previously worked as Developer Relations Engineer at Neo4j. Apart from writing blog posts and creating videos, Mark is dedicated to the developer experience, simplifying the process of getting started by making tweaks to product and documentation. Mark writes about his experiences working with all things data at markhneedham.com. Connect with him on Twitter.
In his NODES 2022 presentation, he exposes a graph model before doing a quick walk-through of it based on his predictions of tennis tournaments in 2022. Watch his talk “Graph Modeling The Shadow Graph”!
GraphGPT, an open source project by Varun Shenoy, converts unstructured natural language into a knowledge graph. Use your OpenAI API key to try the web version or install on your machine from the GitHub repo.
In part one describing a recommendation system, Susmit explored collaborative and content-based filtering and designed the data model.
In this second article, he explains writing cypher queries for loading the data, tracking new orders and implementing the recommendations.
Sixing explains how a cloud-native medical chatbot called Doctor.ai, backed by a Neo4j graph was developed. You can employ either AWS Lex, GPT-3, or Alan AI as the natural language understanding engine.
Bojan Ciric shows how to present a convincing graph solution to executives. He explains how the knowledge graph is the solution to maintaining an enterprise view of data, a key problem of data decentralization.
Yolande Poirier is passionate about technology and developer communities. Her goal is to empower developers and data scientists everywhere to successfully grow their projects. At Neo4j, she runs the advocacy programs, including the Ninja program. Feel free to reach out to her on LinkedIn.