A model to detect potentially fraudulent/abnormal wires of an insurance company: An unsupervised rule-based approach

Fraud prevention/detection is an important function of internal control. Prior literature focused mainly on fraud committed by external parties, such as customers. However, according to a survey by the Association of Certified Fraud Examiners (ACFE 2009), it was noted that employees posed the greatest fraud threat. This study proposes profiling fraud using an unsupervised learning method. The fraud detection model is based on potential fraud/anomaly indicators in the wire transfer payment process of a major insurance company in the United States. Each indicator is assigned an arbitrary score based on its severity. Once an aggregate score is calculated, those wire transfer payments whose total scores are above a threshold will be suggested for investigation. Our contribution is to report what we have learned and to document our findings using fraud/anomaly indicators to detect potential fraud and/or errors on real data from a major insurance company.