Part B: Generate real-time fraud predictions - Amazon Fraud Detector

Part B: Generate real-time fraud predictions

In Amazon Fraud Detector, you create and configure detectors to hold your deployed model and decision logic (that is, rules). In Part B, You create rules for your detector. These rules outcomes (such as to flag an e-commerce transaction if the ML score is too high).

To get a fraud prediction for an event, you send metadata about an online event using the GetEventPrediction API and then synchronously receive a fraud prediction outcome and model score. In Amazon Fraud Detector, you configure fraud prediction logic using these components: event types, detectors, models, rules, variables, and outcomes.

To create your fraud prediction logic in Amazon Fraud Detector, follow the steps in this section.