Computational Intelligence Models for Insurance Fraud Detection: A Review of a Decade of Research

This paper presents a review of the literature on the application of data mining techniques for the detection of insurance fraud. Academic literature were analyzed and classified into three types of insurance fraud (automobile insurance, crop insurance and healthcare insurance) and six classes of data mining techniques (classification, regression, clustering, prediction, outlier detection, and visualization). The findings of this review clearly show that automobile insurance fraud detection have also attracted a great deal of attention in recent years. The main data mining techniques used for insurance fraud detection are logistic models, Decision tree, the Naïve Bayes, and support vector machine.