First Published 7th January 2021
The role of behavioral biometrics must be ascertained as it relies on dynamic data. It’s prowess lies in understanding both legitimate shopper and fraudsters’ journey, writes Ai’s Ritesh Gupta
7th January, 2021
Travel merchant must adjust to novel buying patterns that have emerged in the last nine months or so. If they don’t then they might end up with a set of unsatisfied customers as well as satisfied fraudsters.
Think of evolving purchasing hours, device usage…there are too many variables to consider in today’s shopping environment.
The time of the day when transactions are being conducted has changed as consumers have been working from home. Also, more people are making e-commerce buys on the behalf of others, as indicated by Vipin Surelia, Head of Risk, India & South Asia, Visa, during Ai’s ATPS virtual conference in October. This is also resulting in a shift in fraud pattern, added Surelia.
So in a fast-changing shopping environment and the associated fraudulent activity, if the defense layer can’t adapt swiftly then a merchant is likely to end up with an alarming number of false positives.
How to tackle the issue then?
A major concern, as always, is to ensure a fraudster’s attempt shouldn’t go unnoticed, whereas a genuine customer doesn’t end up being denied to pay for a transaction.
Progress has been made in the arena of dynamic detection and merchants must gear up and incorporate the same in their defense mechanism. Tools must be able to adapt in real-time to ensure the balance between the user experience (UX) and security isn’t disturbed during the booking flow and is rather optimized.
Fraud prevention specialist SecuredTouch’s co-founder, Ran Shulkind, rightly points out that you can’t see what you’re not looking for.
This is where the role of behavioral biometrics comes into play. It uses machine learning in order to adapt and learn from the moment a user session begins.
As for what it is based on, it considers how a shopper’s dynamic behavioral characteristics interact with a device, and is being utilized on both desktop and mobile devices. For a mobile device, the level of pressure exerted to the screen, the angle at which the device is held and the speed of finger movement across the screen are individual to each particular user, and can now be used to accurately identify fraud, according to specialists. A use case: in case a fraudster is committing an account takeover and working on the same manually then the behavioral pattern can be identified. Behavioral data features numerous sensor readings to uncover intricate and nuanced gestures. So the efficacy lies in understanding the actual behavior to gauge subtle differences between a legitimate user and a fraudster. “This technology is designed to identify threats you aren’t looking for,” asserts Shulkind.
Being pragmatic and ready
According to SecuredTouch, behavioral biometrics delivers continuous authentication. This means that data is continuously being collected in real-time. This data is used to optimize machine learning models so as to continually assess legitimate customer journeys. These can be contrasted with behavioral patterns that may be indicative of fraud.
Machine learning’s role in fraud detection can’t be undermined but it shouldn’t be forgotten that it takes time to recalibrate.
A human fraud analyst can take charge and control the situation, ensuring a genuine shopper’s experience isn’t hampered. Merchants must assess how to make the most of a machine learning system with a rules-based approach.
Buy-in or testing for any new project it isn’t a straightforward task. At the same time, it is imperative for travel e-commerce players to have a long-term perspective. Any effort made now to secure accounts or protect customer data will go a long way in protecting the interests of an entity as well as serving the customers when they are ready to travel again.
