Using Kohonen's Self-Organizing Feature Map to Uncover Automobile Bodily Injury Claims Fraud

TLDR

Claims fraud is an increasingly vexing problem confronting the insurance industry. The study applies Kohonen's Self‑Organizing Feature Map to classify automobile bodily injury claims by fraud suspicion. The authors use a SOM trained with feed‑forward neural networks and back‑propagation to validate the fraud‑suspicion classification approach. Experiments show the SOM method outperforms insurance adjusters and investigators in consistency and reliability.

Abstract

Claims fraud is an increasingly vexing problem confronting the insurance industry. In this empirical study, we apply Kohonen's Self-Organizing Feature Map to classify automobile bodily injury (BI) claims by the degree of fraud suspicion. Feed forward neural networks and a back propagation algorithm are used to investigate the validity of the Feature Map approach. Comparative experiments illustrate the potential usefulness of the proposed methodology. We show that this technique performs better than both an insurance adjuster's fraud assessment and an insurance investigator's fraud assessment with respect to consistency and reliability.