Amazon Web Services (AWS) on Tuesday announced the general availability of a fully managed service designed to help customers identify potentially fraudulent online activities.
Leveraging machine learning, the Amazon Fraud Detector can spot payment and identity fraud almost instantly.
Customers are provided with pre-built templates directly in the Amazon Fraud Detector console, where they can also upload historical event data and choose actions to be taken.
The new service, which AWS unveiled last year, is provided without long-term commitment or up-front payments, and customers will be charged for the actual usage only, the company says. The service doesn’t require an infrastructure to manage either.
According to AWS, the service can detect in real time potential fraud related to online payments and identity, fake accounts, abuse of loyalty accounts and promotion codes, and the like.
Customers can train and deploy machine learning models based on historical data of fraudulent and legitimate transactions. The service only requires the provision of two attributes associated with an event, but additional data can be included as well.
Amazon Fraud Detector processes the data, selects an algorithm, and then trains and deploys a model to a fully managed API endpoint.
Fraud risk responses are delivered for all of the activity that is sent to the API, and the customer’s application leverages these responses to determine what action should be taken.
Amazon Fraud Detector also allows for customizations through combining machine learning models built using the new service with those trained using Amazon SageMaker.
AWS has made Amazon Fraud Detector available in the United States (N. Virginia, Ohio, and Oregon), Europe (Ireland), and Asia Pacific (Singapore and Australia – Sydney). The service is expected to roll out to more regions in the coming months.
“By leveraging 20 years of experience detecting fraud coupled with powerful machine learning technology, we’re excited to bring customers Amazon Fraud Detector so they can automatically detect potential fraud, save time and money, and improve customer experiences—with no machine learning experience required,” said Swami Sivasubramanian, VP of Amazon Machine Learning at AWS.