Graph neural networks are turning hidden fraud into visible networks
Graph neural networks are reshaping how enterprises hunt for fraud, moving detection beyond isolated transactions to reveal entire hidden networks of bad actors. As AI adoption accelerates, organizations are discovering that the real breakthrough isn’t just faster models — it’s a data structure built to expose relationships that traditional systems miss. That shift is playing […] The post Graph…
Graph neural networks are revolutionizing enterprise fraud detection by uncovering hidden networks of fraudulent activities. Thomas Luu, director of global product security at Gilead Sciences Inc., explained to theCUBE's John Furrier at the Neo4j GraphTalk event how his team leverages graph neural networks to detect fraudulent activities in the pharmaceutical industry.
Prior to adopting graph technology, Luu's team manually compared disparate data sets, a process that was limited by human expertise and scalability. The new three-layer detection model combines rules-based logic, traditional machine learning, and graph neural networks to uncover relationship-based fraud schemes. Graph neural networks reveal hidden relationships between entities that traditional systems fail to detect, allowing Luu's team to group main fraud actors with their supporting accomplices.
This capability not only improves detection speed but also provides an intuitive visual representation of relationships for non-technical investigators.
Written by urgent.news from SiliconANGLE's reporting — not their text. Machine-written — it may contain errors, so check the original before relying on it.
