A top 10 global bank implemented Ayasdi AML to work with the bank’s existing TMS and improve the results of existing AML systems. Ayasdi AML reduced false positives by more than 20% without missing a single SAR.
Intelligent segmentation is the crucial first step for a Transaction Monitoring System (TMS) to accurately detect suspicious events. Ayasdi AML’s auto feature engineering identifies attributes within data that contain signals and derives new attributes that accelerate and enhance its segmentation.
Ayasdi AML ingests the greatest volume and variety of data available—about customers and transactions—and then applies objective machine learning to create the most refined and up-to-date segments possible. The crucial difference is that Ayasdi AML assigns—and reassigns—customers to segments based on their actual behavior, revealed in their real transactions and true inter-relationships, over time.
By analyzing customer transactions on a daily basis, Ayasdi AML automatically lists customers showing a significant change in behavior over time. This gives an investigator the ability to understand multiple viewpoints and flag deviations deemed significant:
Ayasdi AML surfaces far fewer—and far more valuable—events for your investigators to consider. That’s because our machine learning algorithms get continuously smarter with inputs from your subject matter experts about which patterns matter most. Investigators can quickly classify events as alerts or dismiss them requiring no further action.
With these capabilities come clear and comprehensible context so investigators can quickly and effectively inspect an alert and pass SARs to regulators. The system provides clear documentation in the form of an audit trail of why a customer’s behavior qualifies as suspicious so investigator’s decision is based on facts and defensible in the long run.