If there’s one thing that unites the frustration levels of insurers hoping to fight claims fraud, it’s the human factor: Behind every great fraudulent claim is a human being—or maybe a group of human beings being “all things human.” I think in this case we can define “human” as being far from perfect, maybe being tempted, or valuing greed, or thinking they can outsmart the authorities.
For insurers, this human factor creates a never-ending “cat and mouse” game of trying to second-guess the fraudster. And from a business perspective, this can be an exhaustive, expensive endeavor.
The latest tools and techniques seem to give the cat the advantage, because fraud discovery no longer hinges on when the fraudster is most human (makes a mistake and gets caught); instead, it gives equal play to data that can be used to back a proactive strategy to predict and prevent as well as to prove its case.
Using data analytics to improve claims outcomes isn’t new. For years, insurers have relied on data stores and models to help them determine which claims are valid, which bills presented were for legitimate repairs or services rendered, etc.
Today, however, entity analytics (EA) has entered the discipline. EA can be considered both a science and a methodical process that is designed to detect like and related entities across disparate data stores, applying advanced techniques to old and new data to ascertain connections that would otherwise not be obvious.
EA is already being applied to more than just financial services and insurance enterprises, being used in perpetual employee vetting, insider thread detection, watch-listing programs, critical infrastructure protection, counter terrorism, counter intelligence, identity fraud solutions and more.
Rick Hoehne, IBM’s global leader of business analytics and optimization and Jeff Huth, program director, production manager for IBM’s entity analytics product portfolio, provided attendees of a recent webinar with information on how EA can improve claims fraud detection and prevention outcomes that apply to almost all types of insurance fraud, from organized events such as staged accidents and third-party fraud, to opportunistic fraud such as exaggerated losses.
One of the hallmarks of the EA offering is, in simple terms, its ability to accelerate the claims investigation process by learning who is who, who is related to who (and what), and who is doing what.
Key to this function, said the presenters, is incremental context accumulation, which amounts to collecting real-time, perpetually updated data that detects similar records, related records and activity patterns.
The notion of building context accumulation entity information automatically sounds like a tall order for any claims organization, and the presenters admitted that insurers could be doing much more to take advantage of this technology.
Weighing the cost/benefit, there’s no reason not to, since the technology is available from more than one reputable source. Since 2007 IBM has systematically built out its analytics offerings with the acquisition of Cognos and SPSS, which grandfathered EA into to its analytics suite. No doubt IBM and its EA-charged InfoSphere suite will go head-to-head with privately owned SAS, the largest analytics software vendor of its kind (which also offers entity resolution products), for insurance industry share of market. The creation of a competitive landscape should bode well for insurers shopping for this type of system.
Ironically, in spite of the lofty technology promise, the idea of EA continues to comprise “human” qualities. In describing the importance of entity analytics to a variety of vertical markets, Jeff Jonas, IBM fellow and chief scientist of the IBM entity analytics group, makes a compelling argument for its adoption, telling his blog’s audience that “When organizations can process arriving observations [transactions/records] for relevance … organizations will be more competitive, or might even, for the first time, seem to be ‘awake.’”
Pat Speer is an editorial consultant for Insurance Networking News.
Readers are encouraged to respond to Pat by using the “Add Your Comments” box below. Shealso can be reached at firstname.lastname@example.org.
This blog was exclusively written for Insurance Networking News. It may not be reposted or reused without permission from Insurance Networking News.
Register or login for access to this item and much more
All Digital Insurance content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access