p1b1sh04so1auprh8onbpv28c86.jpg
In a recent whitepaper, “Innovations in Using Social Media to Fight Insurance Fraud, Improve Service,” insurer QBE says that “Social media offers nearly endless possibilities for fraud investigators… But the massive volume of social media content is also the biggest challenge.” To help, the company outlines five technologies that insurers can leverage to parse, link and analyze social content. “The most recent and best service approaches place considerable emphasis on interconnectivity between online information platforms such as social media, traditional websites, and public record databases,” QBE says.
p1b1sh04soblucc41ri11jnvnf7.jpg

Text-mining software

Already used to help investigators examine internal claims data such as adjuster notes, emails, service calls and interview records, QBE notes this can be used to search social media sites including Facebook, LinkedIn, YouTube and Twitter.
p1b1sh04spoen1vng1ubl13il1lmg8.jpg

Social CRM

These tools search millions of social media posts and platforms to find activity related to recent claims. For example, it can be used to identify connections between people in an accident, or find out about plans for properties that happened in advance of claims on them.
p1b1sh04sp6g8kra1j5s1p121qd89.jpg

Predictive modeling

This “aims to spot suspicious activities as early in the claims cycle as possible, since preventing fraud is the key objective and recovery is significantly more difficult and expensive once a fraudulent claim has been paid out,” QBE writes.
p1b1sh04spqld1q0m1mkskqn1qeua.jpg

Cross-leveraged data sources

These sources, such as telemetry, can be coordinated with social media information, QBE says. Motorists with telematics devices installed can also be used to pinpoint the truth behind accident claims. “For instance, fraudsters claiming they were rear-ended may be confronted by data showing their victim’s car was standing still at the time of impact,” the company writes.
p1b1sh04sqp5st1ihgr1li0g5tb.jpg

Social network analysis

This technology sifts through large amounts of data to uncover hidden relationships among people, places, locations, accounts or virtually any other type of entity. For example, it can be used to identify when multiple claims all involve the same vehicle or originate from a single household.