buguroo’s behavioral biometrics tool closes 170 mule accounts in one month
buguroo has upgraded its bugFraud 4.0 solution that prevents banking fraud with a brand new capability, which leverages behavioral biometrics and context information to identify behavioral patterns through profiling and to predict future fraud schemes in online channels.
“Online fraud attacks are continuously evolving and becoming more sophisticated in their attempt to bypass anti-fraud solutions,” said buguroo CTO Jose Carlos Corrales in a prepared statement. “Traditional fraud prevention technologies are focused on identifying the attack itself and are proving ineffective against this threat; the solution also requires fraudster identification to check fraud at its roots.”
In a Q&A email exchange with Biometric Update, buguroo’s Vice President for EMEA, Tim Ayling, shared information about the company’s motive to roll out Fraudster Hunter, deployment details and how its customers have already benefited from the upgrade.
Was the motivation behind Fraudster Hunter to improve fraud prevention or were customers asking for such a feature in the solution?
As fraudsters continuously adapt and find ways to circumvent banks’ anti-fraud practices, it is increasingly important that fully comprehensive anti-fraud solutions can detect which fraudsters are active in a bank’s infrastructure, instead of just deciding whether a user is legitimate or not.
The behavioral biometrics functionality helps banks confront some major fraud challenges, such as new account fraud, mule accounts, and the use of synthetic identities. Traditional anti-fraud methods have proven ineffective in detecting this type of attacks, but Fraudster Hunter makes it significantly easier.
Banks can now directly identify who fraudsters are, instead of just reacting to new and different types of fraud attacks. With Fraudster Hunter, they are one step ahead because they cut off fraud at its root.
How does the solution leverage behavioral biometrics to identify fraudsters and detect patterns?
Similar to how the bugFraud solution analyses thousands of user behavior parameters to confirm a customer’s identity, such as the trajectory on which a user typically moves the mouse or the speed and rhythm with which they type, Fraudster Hunter is an added functionality that applies the same analysis process to identify fraudsters.
Not only does Fraudster Hunter recognize typical fraudulent behavior, it can then visualize graphically, or ‘map’, the relationships between events during each online session. Then we can drill down into the more complex interrelationships between organizations, people and transactions, and use this information alongside our deep learning technology to find connections between known fraudulent applications and other, potentially fraudulent applications.
By applying this method to analyze fraudster behavior, the bank can draw conclusions about the unique modus operandi of a known fraudster attack and consequently identify more questionable patterns and expose fraudulent accounts.
Have bank analysts noticed any progress since integrating Fraudster Hunter?
Fraudster Hunter has already played a valuable part in helping the police identify a network of criminals. Officers already knew that the criminals it was investigating had accounts at a particular bank, after their victims had paid money directly into these specific accounts.
Once it integrated Fraudster Hunter, the bank easily identified individual fraudsters by analyzing their typical interactions with its services. In fact, link analysis detected a number of additional accounts controlled by the same fraud ring, which led to the shutdown of more than 170 mule accounts in just one month.
In January, buguroo upgraded the solution with a number of new features, including environment analysis, which evaluates geolocation, device and network data and reputation to verify identity, behavioral biometrics for onboarding control and contextualized detection techniques.