AWS announces phone number enrichments for Amazon Fraud Detector Models
We are excited to announce the launch of phone number enrichments for Amazon Fraud Detector machine learning (ML) models. Amazon Fraud Detector (AFD) is a fully managed service that makes it easy 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 from AFD automatically identifies potentially fraudulent activity in milliseconds—with no ML expertise required.
We are excited to announce the launch of phone number enrichments for Amazon Fraud Detector machine learning (ML) models. Amazon Fraud Detector (AFD) is a fully managed service that makes it easy 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 from AFD 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 address and Bank Identification (BIN) number of payment instruments with data such as the geolocation of the IP address or the issuing bank for a credit card. Augmenting users’ data with such enrichments ensures best-in-class performance from AFD models. Starting today, Amazon Fraud Detector now enriches phone number data with additional information such as geolocation, and the original carrier. This new enrichment boosts performance for models that use phone number, enabling these models to capture up to 16% more fraud at a 4% false positive rate.
Phone number enrichments are automatically enabled for AFD’s Online Fraud Insights (OFI) and Transaction Fraud Insights (TFI) model types in all regions where AFD is available. AFD customers can make use of this new enrichment by retraining their AFD models that use phone number as one of the event variables. For additional details, see our documentation page.