Claims Fraud Detection using
Manual claims assessment is a big expense for insurance firms. A VC-funded software company set out to devise software that scored claims based on likelihood of fraud to focus manual review appropriately using multiple data sources including location data.
Gravitate complemented the client’s internal data science and technology team by:
Bringing in additional data sources including via a web scraping capability.
Advising on model building and algorithm development.
Providing additional support to combine location data with other data sources, and visualize in a python pipeline.
A fully-operational solution was developed in less time vs. the initial estimate. The client has engaged with a major insurance company in its market.