One of the use cases for Big Data that may push the limits of just how big that data can get is in fraud prevention. Not surprisingly, financial and insurance firms, with potentially the most to lose or gain, are moving past their initial forays into big data fraud prevention and detection projects, and exploring more global approaches to match the global reach of fraudsters.
One U.S. insurance firm, for example, suspected there was more behind seemingly unrelated but fishy insurance claims, but couldn’t find the connection. With the services of LexisNexis Risk Solutions, however, the claims were revealed to have come from related parties that were colluding to defraud the firm. According to the account at ITWeb, LexisNexis used data on “social” connections to expose the fraud.
VP of Information Security at LexisNexis Risk Solutions, Flavio Villanustre, says in the article that the reality is that “criminals are getting smarter.” Just as your firm is performing risk assessments, “they are assessing the risks and realizing that there is a lot less risk involved in doing cyber crime than there is in going out and physically stealing something.” Maybe that seems a little like stating the obvious, but part of this “smarter criminals” trend is that big data can work for anyone, not just the good guys, and not just the big guys.
In another fraud-rich environment, medical insurers are increasingly advised to approach data collection as a story. Just a few months ago, when IT Business Edge’s Loraine Lawson wrote about the data storytelling concept, it was described as the “domain of research institutes and Web 2.0 companies.” Now, insurance companies, or “payers,” writes Dina Overland at FierceHealthPayer, are focused on predictive analytics as a group, and “structuring the data like a reader might chart pages of a complex narrative in a book.” And insurers, she quotes Mark Golberg, general manager of the provider, payer and ACO sectors at Recombinant by Deloitte, should expand the data sets by taking “advantage of certain social media sites like StreetRx and Twitter for pricing and doctor information.”
In financial services, where the total damage from credit card fraud has reached some $5.55 billion, banks and financial services are deep into aggregating data from multiple sources, including social media, to profile customers’ behavior and detect fraud, writes Cognizant’s Karthik Krishnamurthy at Banktech.com. And by layering this clustering of data from social sources with agile systems that learn from each other, the percentage of false positives in fraud detection is going down–a key improvement metric for customer service.
It’s most definitely a race against the large-scale fraudsters, writes Krishnamurthy:
“… big data will help deal with the globalization of fraud itself. On a global scale criminals are developing fraud tactics and scenarios driven by data and analytics. They use data to probe for weaknesses and monitor the ‘success’ of fraud programs they initiate. As computing resources become cheaper and faster in an internet world where a ‘location’ is very much a virtual presence, criminal enterprises can move operations in a borderless digital world.”