OREANDA-NEWS. September 24, 2015. “Insurance companies have always had certain measures in place to detect possible fraud early in the claims process,” says Thomas Djursoe. “However, the unceasing increase in the complexity of insurance frauds has exposed the limitations of traditional fraud-detection systems, such as internal audits, whistle-blower hotlines to report fraud, and software that flags anomalies based on a pre-defined set of rules.” Excerpts:

“By analyzing customer data alongside behavior, insurance companies now have extra tools against fraud that otherwise would be hard to detect using more traditional methods.

Some insurance companies today are offering drivers a tracking device (the Blackbox) to install in their vehicles in order to gain insights into the motorist’s driving patterns or, in the case of theft, track the vehicle. The sensors in this tracking device monitor the driver’s behavior behind the wheel, if there is speeding or any harsh braking involved or sudden turns are taken, and so on. Therefore, a much more effective risk assessment of the driver profile is now given to the insurance companies than beforehand.

All this data, or “Car Halos”, in this instance, surrounds each driver and can be tapped into and analyzed. Using analytics and real-time information generated by the telematics and sensors installed in the car, insurance companies get a clear picture of the risks involved in the driving and thereby more easily identify fraudulent claims.

Social media and network analytics can help to identify proximities and relationships among people, groups, organizations and related systems, helping to detect and investigate fraud.

Text analytics helps companies gain critical insights from large volumes of unstructured data, such as notes the adjuster takes when interviewing a claimant, during the initial contact with the insurance company regarding a claim or incident, e-mail correspondence and accident descriptions, which often consist of short or incomplete sentences, misspelled words and abbreviations.

Link analysis provides the larger picture for a claim. For instance, in the case of a car accident involving multiple claimants, link analysis can use claimants’ addresses, phone numbers, vehicle number, etc. to discover the links among the claimants, the clinics where the claimants were treated and repair shop they used, thus leading investigators to rings of professional fraudsters.

In order for insurance companies to apply analytics on the large amount of data on their customers, they need an efficient model and approach to enterprise-wide data management.

Ultimately, insurers who send a strong message about their commitment to fighting fraud will be the ones securing the trust of existing and future customers and investors as well as reducing premiums as a result of fewer fraudulent payments.”