E-commerce payments technology provider Stripe is launching a fraud prevention tool for security professionals at large organizations.
Radar for Fraud Teams is designed to help security teams review payments faster and monitor information related to those transactions that have been processed at the business.
The fraud logic analytics will help businesses establish custom rules for transactions and receive real-time feedback regarding e-commerce transaction data.
“Stripe’s machine learning models are now trained on hundreds of billions of individual data points drawn from the Stripe network," Michael Manapat, engineering manager for Radar and machine learning at Stripe, said in a Wednesday press release.
"We’ve used these data points to update our fraud models, helping businesses on Stripe more accurately identify fraudsters and reduce fraud rates while still keeping payment acceptance rates high," Manapat said.
Stripe says the original version of Radar prevented $4 billion in attempted fraud in 2017 by learning from the transactions processed on the Stripe network for hundreds of thousands of businesses.
Radar for Fraud Teams represents an update to the software's 2016 machine learning models and adds hundreds of new signals that distinguish legitimate customers from fraudsters, Stripe said.
Proxy detection provides a signal incorporated into Radar that measures round-trip time between Stripe and a potential fraudster's browser. Such detection helps fraud teams pinpoint if a fraudster is using a proxy or a virtual private network.
At the same time, Radar constantly evaluates patterns that are unique to a user's business. It will also update and retrain its own models on a daily basis, while evaluating each transaction moving through the network.
San Francisco-based Stripe provides its technology platform for developers building internet businesses. Its software focuses on payment acceptance, fraud prevention, and various billing/payments models.