Hopsworks provides a JDBC connector to enable Neo4j tables to be an external feature group in Hopsworks. Neo4j tables can also be a source for feature pipelines, with features stored in Hopsworks’ Online/Offline stores. Neo4j tables can also be sinks for predictions or logs produced by inference pipelines.
You can leverage data in Neo4j to build real-time features in Hopsworks. You can also centrally manage features stored in Neo4j on Hopsworks, enabling them to be joined with features from other data sources. You can easily use Neo4j tables as data sources for feature pipelines.