While BI solutions can mine big data to flag known threats, they cannot easily detect new ones. Graph data enables efficient discovery of emerging patterns, but at scale, performance suffers.
In this session, Weidong will demonstrate efficient detection of threat patterns using a tiered architecture that integrates real-time data with graph-enabled visualization.
With a tiered system, risk-related events reduce to a smaller set of graph patterns or basic observations. Through visualization and exploration of patterns stored in Neo4j, you can define higher-level threat patterns. Real-time connection to the data store allows you to expand upon existing patterns. You then proactively implement automated alerts to prevent damage.
#Intermediate #Visualization # Python #GraphDataScience #Database #FraudDetection #Neo4k