Amazon Web Services (AWS) has announced the general availability of its machine learning-based fraud detection service.
Amazon Fraud Detector is a fully managed service touted as making it easy to quickly identify potentially fraudulent online activities, such as online payment and identity fraud, the creation of fake accounts, and loyalty account and promotion code abuse “in milliseconds”.
To use the service, customers can select a pre-built machine learning model template; upload historical event data of both fraudulent and legitimate transactions to build, train, and deploy machine learning models; and create decision logic to assign outcomes to the predictions.
“Customers of all sizes and across all industries have told us they spend a lot of time and effort trying to decrease the amount of fraud occurring on their websites and applications,” Amazon machine learning VP Swami Sivasubramanian said.
“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.”
Based upon the type of fraud customers want to predict, Amazon Fraud Detector will pre-process the data, select an algorithm, and train a model.
Amazon Fraud Detector trains and deploys a model to a fully managed, private API endpoint. Customers can send new activity to the API, such as signups or new purchases, to receive a fraud risk response, which includes a fraud risk score. Based on the report, a customer’s application can determine the right course of action, for example, to accept a purchase, or pass it to a human for review, AWS explained.
See also: How Amazon Web Services runs security at a global scale
Announcing the service at AWS re:Invent in December, CEO Andy Jassy said at the time that his company is always looking at what it can offer, including within its in-house capabilities, to give value to its customers.
“We have a number of services that we’ve done at scale at Amazon that customers have asked for as a service,” he said, pointing to Amazon Lex, Amazon Personalize, and Amazon Forecast as recent examples.
“We’ve been doing fraud detection for over 20 years.”
Jassy said machine learning is “unbelievably helpful” but that it is hard for most companies to use machine learning for fraud detection.
“You send us the transaction data … email addresses, IPs, perhaps phones numbers, along with transactions that are fraudulent and those which are legitimate and then we take that data along with the algorithms we’ve built … and we build a unique model for you,” he added.
“It’s then exposed via an API.
“A completely different way to manage fraud with machine learning.”
Amazon Fraud Detector is available today in US East (Northern Virginia), US East (Ohio), US West (Oregon), EU (Ireland), Asia Pacific (Singapore), and Asia Pacific (Sydney) regions, with additional regions to be added in the coming months.
