Visa’s chief enterprise risk officer, Ellen Richey, says “you see the criminal capability evolving on the technology side.” She gives CIO Journal an inside look at how the company has used Big Data to make its network more secure...
The approach Visa takes in identifying fraud is grounded in 16 different predictive models and allows for new independent variables to be added to the model. This improves accuracy while alowing the models to be kept up to date. Here's an excerpt from the WJS Article:
"The new analytic engine can study as many as 500 aspects of a transaction at once. That’s a sharp improvement from 2005, when the company’s previous analytic engine could study only 40 aspects at once. And instead of using just one analytic model, as it did in 2005, Visa now operates 16 models, covering different segments of its market, such as geographic regions."
The article also states that the analytics engine has the card number and not the personal information about the transaction - likley stored in a different system. I wonder if Visa, at some point in the process, also takes the fraud transactions and analyzes them visually to identify connections and linkages based on address, other geographic identifiers, 3rd party data, employer data and more? Are two or more of the fraud cases in some way connected? Does this represent a ring of activity presening higher risk to merchants, customers and Visa?
The tools on the market to do this work are expanding. The data used to analyze this activity (including unstructured data) is being stored in databases that allow for the visual analysis of big data. Graph databases replete with underlying intelligence extracted from text that identify people, places and events can be used to extend the type of analysis that Visa is doing and prioritize investigations. Through more efficient allocation of investigation resources, fraud prevention can jump to a higher level.