Skip to content
On-premise users: click in-app to access the full platform documentation for your version of DataRobot.
DataRobot API resources > AI accelerators > Use cases and horizontal approaches > Build a fraud detection pipeline with Neo4j

Build a fraud detection pipeline with Neo4j

Access this AI accelerator on GitHub

This accelerator demonstrates how to build a fraud detection pipeline using Neo4j and DataRobot. Use Neo4j to store and query a knowledge graph of clients, loans, addresses, and more. Then, use DataRobot to build a predictive model with graph-based features. The accelerator contains multiple notebooks. The first notebook walks through installing a Neo4j 4.4.11 instance, loading a Neo4j database, and uploading a dump file with the CLI. The second notebook outlines how to extract graph data into training and holdout CSVs, upload training data to DataRobot, and build a classification model for scoring.