Just as the operations of criminals perpetrating fraud have become more complex and sophisticated, our efforts to fight them have also advanced. Some of FICO's latest work has seen the introduction of self-learning models that allow the system to adapt more quickly to customer behaviour and reliably catching fast-changing fraud patterns.
Most recently, we have also started looking at location-based ways to catch fraud. FICO is currently developing a service,which checks if a cardholder's registered mobile phone is near the ATM or point-of-sale terminal where the card is being used. Based on the proximity, a cardholder account can be blocked or cleared for further activity. This in tandem with predictive analytics will be incredibly powerful against fraudsters.
How can big data and analytics help enterprises/financial institutions to combat fraud?
Predictive analytics is a technology that helps make better decisions based on modelling techniques and comparing them to large amounts of live streaming data. Analytics software can look at historical activity to determine whether current activities are suspicious and then flag it for action by a bank.
For fraud, real-time detection has become very critical. Analytics can help detect fraud on payment card authorisation and electronic payment transactions for a bank in a fraction of a second.
Looking at behaviours and patterns is key as banking customers will all have a profile of habits. They may be withdrawing money only from ATMs near work or home, shopping mostly on weekends or purchasing certain categories of goods. Behaviour that is a clear departure from such routines would be a red flag for potential fraud.
What else should enterprises do to prevent themselves from falling prey to fraud?
Fighting fraud is a constant battle. What enterprises need to do is build a multi-layered defense model. They also need to secure their data against a breech and follow data protection protocols.
As mentioned, fraudsters are opportunistic and will target those that are easy targets, so having some more sophisticated protection is a competitive advantage.
For banks this might include fraud analytics that use "Behaviour Sorted Lists". This technology helps distinguish between "in-pattern" normal customer spending and "out-of-pattern" suspicious transaction activity. It enables faster fraud detection at much lower false positive rates (declines on legitimate transactions).
These "Behaviour Sorted Lists" are updated with each transaction so that the customers' patterns of favourites evolve over time. The more frequent entries appear with greater recurrence and are ranked at the top of the list, while less frequent entries fall away and are replaced with new entries.
Such inventive self-learning systems are improving the way, banks and even other enterprises are mitigating fraud.
Even though prevention is said to be better than cure, enterprises - especially financial institutions - need to have the mentality that they will be hacked one day. How should they prepare themselves to mitigate the impact of fraud when it happens to them?
Fraud is always a risk, and even the most advanced banks may be vulnerable to fraud attacks. If a breech does occur, how banks are handling the fraud event has become a defining moment in customer satisfaction and brand loyalty.
Sign up for Computerworld eNewsletters.