Abstract: Fraud is a multi-billion dollar problem facing everyone from banks to credit card issuers to payment processors to merchants. Detecting and preventing fraud has traditionally been a challenge for these types of businesses, which rely on a clunky combination of heuristic-based rule engines and home-grown analytics solutions. In this talk, Danny D. Leybzon will be presenting a new methodology for anomaly detection and analysis that can be applied to everything from fraud detection to factory accident prevention. This system uses a combination of the Imply analytics platform (built on top of the open-source Apache Druid) and the anomaly detection system Sherlock (built on top of the open-source Yahoo EGADS). It leverages both human and machine expertise, allowing both actors to play to their strengths while offsetting each other’s weaknesses.
Bio: Danny lives and breathes data science. He is constantly finding connections and similarities between the world of data science and the world around us. Danny loves what he does. The feeling of solving interesting problems is exhilarating to Danny, and he loves seeing the impact of machine learning and advanced analytics to improve society and lives.