Posted On: Feb 8, 2022
We are excited to announce the launch of a new geolocation enrichment feature for Amazon Fraud Detector machine learning (ML) models that automatically calculates the distance between the IP address, billing address, and shipping address provided for an event. This helps you to prevent more fraud, particularly when a user attempts to create an account with someone else’s information or make a transaction with someone else’s credit card.
Amazon Fraud Detector is a fully managed service that makes it easier to identify potentially fraudulent online activities, such as the creation of fake accounts or online payment fraud. Using ML under the hood and based on over 20 years of fraud detection expertise, Amazon Fraud Detector automatically identifies potentially fraudulent activity in milliseconds—with no ML expertise required. As part of the model training process, Amazon Fraud Detector enriches raw data elements like IP addresses, bank identification numbers (BINs), and phone numbers to create dozens of additional inputs to your fraud detection model.
Starting today, Amazon Fraud Detector automatically calculates the physical distance between the IP address, billing address, and shipping address you provide for an event. The calculated distances are then used as inputs to your fraud detection model. This new enrichment boosts performance for models that use these variables, enabling these models to capture up to 10% more fraud when accepting a 3% false positive rate
Geolocation enrichment is automatically enabled for all models in all regions where Amazon Fraud Detector is available. Amazon Fraud Detector customers can make use of this new enrichment by retraining their models that use IP Address as well as Shipping Address and/or Billing Address variables. For additional details, see our documentation page.