 Auto insurance fraud detection: Leveraging cost sensitive and insensitive algorithms for comprehensive analysisMeryem Yankol Schalck Insurance: Mathematics and Economics, 2025, vol. 122, issue C, 44-60 Abstract: As technology and the economy continue to grow, fraud has a significant negative impact on business and society, and insurance fraud remains an important issue, posing challenges in both detection and prevention. This article provides a direct cost-sensitive learning approaches on enhancing traditional motor insurance fraud detection by leveraging real-world data sets. In this approach, the results are obtained by using the information available at the opening of the claim, FNOL. The data set (FNOL) contains numerical, categorical, and textual variables. The results show that machine learning techniques perform better statistically and can also be more effective than standard approaches in reducing fraud-related costs. Extreme Gradient Boosting (XGB) outperforms both cost-sensitive and cost-insensitive approaches based on performance measures. Our study indicates that a cost-sensitive strategy delivers greater financial benefits than a cost-insensitive approach. Keywords: Fraud detection; Automobile insurance; Cost sensitive and insensitive algorithms, Natural language processing (search for similar items in EconPapers) Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:eee:insuma:v:122:y:2025:i:c:p:44-60 DOI: 10.1016/j.insmatheco.2025.02.001 Access Statistics for this article Insurance: Mathematics and Economics is currently edited by R. Kaas, Hansjoerg Albrecher, M. J. Goovaerts and E. S. W. Shiu More articles in Insurance: Mathematics and Economics from Elsevier |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.