Announcing: Graph-Native Machine Learning in Neo4j!

We're excited to announce that graph-native machine learning (ML) is now available in Neo4j, with the release of Graph Data Science Library version 1.4. This library includes graph embeddings and an ML model catalog, enabling users to create representations of their graph and make predictions within Neo4j. The graph embedding algorithms transform the topology and features of a graph into fixed-length vectors that uniquely represent each node. Three embedding options are available: Node2Vec, GraphSAGE, and FastRP, which offer varying levels of performance and accuracy. The ML model catalog stores and references predictive models, allowing users to easily apply them to their data. A real-world example of knowledge graph completion for drug discovery demonstrates the potential of Neo4j's capabilities in this area. With this release, Neo4j aims to democratize advanced graph-based ML techniques, making them accessible to enterprises beyond leading Big Tech companies.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.