Research on Detecting Auto Insurance Fraud

Abstract

With the expansion of the insurance sector has come an increasing concern about insurance fraud. This issue has grown in prominence over time, necessitating the implementation of appropriate countermeasures. In response, this work conducts a thorough analysis of the available literature, with an emphasis on the incorporation of machine learning approaches for detecting fraudulent actions in the field of automobile insurance. By delving into the intricacies of the vehicle insurance fraud landscape, the research investigates the strategic deployment of expert systems and machine learning models to enhance fraud identification processes. Drawing from the wealth of high-quality auto insurance data spanning the last five years, this work explores the implementation of diverse machine learning algorithms, including Logistic Regression, Decision Tree, and Discriminative Analysis, in order to construct comprehensive predictive models. After the modeling was completed, comprehensive testing and analysis were conducted, and the results showed that the logistic regression model had the highest accuracy among these three models. Finally, the future research direction of automobile insurance fraud detection technology is discussed, and how to reasonably prevent automobile insurance fraud is proposed.