Abstract

Fraudulent insurance claims pose a significant financial burden on the vehicle insurance industry, leading to increased premiums for honest customers and substantial losses for insurers. Traditional manual methods for detecting fraudulent claims are inefficient, time-consuming, and prone to errors. This study addresses these challenges by applying and evaluating multiple machine learning techniques to accurately distinguish between genuine and fraudulent vehicle insurance claims. The research specifically aims to characterize the nature of fraudulent vehicle insurance claims, identify key features relevant for model training, assess the performance of various classifiers on both balanced and unbalanced datasets, and develop a web-based system that automates the classification process using the optimal model. Experimental results demonstrate that ensemble methods, particularly AdaBoost and Extreme Gradient Boosting (XGBoost), outperform other classifiers, achieving a classification accuracy of 84.5%. Logistic Regression shows the poorest performance, while Artificial Neural Networks (ANN) perform better with unbalanced data but degrade with balanced data. Additionally, model scalability remains limited to smaller datasets for all evaluated classifiers. The study’s outcomes provide a practical machine learning-driven framework to enhance fraud detection accuracy and processing efficiency, supporting insurers in mitigating losses and improving risk management.

References

• Gill, K. M., Woolley, A., & Gill, M. (2005). Insurance Fraud: The Business as a Cictim? In Palgrave Macmillan UK eBooks (pp. 73–82). https://doi.org/10.1007/978-1-349-23551-3_6
• Association of Kenya Insurers, (2023). Insurance Industry Market Report 2023. Available at: https://www.akinsure.com/content/uploads/documents/Insurance_Industry_Market_Report_2023.pdf?t=0838
• Insurance Regulatory Authority (2023). Insurance Industry Annual Report. Available at: https://libraryir.parliament.go.ke/items/ba83e0b3-b7a2-4583-b671-c022e6da6ec5
• Subudhi, S., & Panigrahi, S. (2018b). Effect of Class Imbalanceness in Detecting Automobile Insurance Fraud. 2018 2nd International Conference on Data Science and Business Analytics (ICDSBA), 528–531. https://doi.org/10.1109/icdsba.2018.00104
• Caruana, M. A., & Grech, L. (2021). Automobile Insurance Fraud Detection. Communications in Statistics Case Studies Data Analysis and Applications, 7(4), 520–535. https://doi.org/10.1080/23737484.2021.1986169
• Viaene, S., & Dedene, G. (2004). Insurance Fraud: Issues and Challenges. The Geneva Papers on Risk and Insurance Issues and Practice, 29(2), 313–333. https://doi.org/10.1111/j.1468-0440.2004.00290.x
• Moon, H., Pu, Y., & Ceglia, C. (2019). A Predictive Modeling for Detecting Fraudulent Automobile Insurance Claims. Theoretical Economics Letters, 09(06), 1886–1900. https://doi.org/10.4236/tel.2019.96120
• Owusu-Oware, E., Effah, J., & Boateng, R., (2018). Biometric Technology for Fighting Fraud in National Health Insurance: Ghana’s Experience. Americas Conference on Information Systems.
• Dull, R. (2014). What Gets Monitored Gets Detected. Journal of Accountancy, Feature Fraud/Technology. http://www.journalofaccountancy.com/issues/2014/feb/20137694.html.
• Burri, R.D., Burri, R., Bojja, R.R., & Buruga, S.R. (2019). Insurance Claim Analysis Using Machine Learning Algorithms. International Journal of Innovative Technology and Exploring Engineering 5(6), Special Issue 4, pp.577-582.
• Dhieb, N., Ghazzai, H., Besbes, H., & Massoud, Y. (2019). Extreme Gradient Boosting Machine Learning Algorithm For Safe Auto Insurance Operations. 2019 IEEE International Conference on Vehicular Electronics and Safety (ICVES), 1–5. https://doi.org/10.1109/icves.2019.8906396.
• Awoyemi, J.O., Adetunmbi, A.O., & Oluwadare, S.A., (2017). Credit Card Fraud Detection Using Machine Learning Techniques: A Comparative Analysis. 2017 International Conference on Computing Networking and Informatics (ICCNI), 1–9. https://doi.org/10.1109/iccni.2017.8123782.
• Gedela, B., & Karthikeyan, P. R. (2022). Credit Card Fraud Detection using AdaBoost Algorithm in Comparison with Various Machine Learning Algorithms to Measure Accuracy, Sensitivity, Specificity, Precision and F-score. 2022 International Conference on Business Analytics for Technology and Security (ICBATS), 1–6. https://doi.org/10.1109/icbats54253.2022.9759022.
• Mishra, A. (2021). Fraud Detection: A study of AdaBoost Classifier and K-Means Clustering. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3789879
• Wilson, J.H. (2009). An Analytical Approach To Detecting Insurance Fraud Using Logistic Regression. In Journal of Finance and Accountancy. 1–3. https://www.aabri.com/manuscripts/08103.pdf
• Pandey, P., (2019). Machine Learning Data Preprocessing: Concepts. Available at: https://towardsdatascience.com/data-preprocessing-concepts-fa946d11c825.
• Bolikulov, F., Nasimov, R., Rashidov, A., Akhmedov, F., & Cho, Y. (2024). Effective Methods of Categorical Data Encoding for Artificial Intelligence Algorithms. Mathematics, 12(16), 2553. https://doi.org/10.3390/math12162553
• Parab, R., (2020). Performance Evaluation Metrics for Machine Learning Models with Python Code. Available at: https://medium.com/swlh/performance-evaluation-metrics-for-machine-learning-models-ad0dd480d5af.
• Patil, S., and Lokesha, V., (2022). Live Twitter Sentiment Analysis Using Streamlit Framework. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4119949
• Tongesai, M., Mbizo, G., & Zvarevashe, K. (2022). Insurance Fraud Detection using Machine Learning. 2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT), 3, 1–6. https://doi.org/10.1109/zcict55726.2022.10046034
• Jalali, B., (2020). Detecting Fraudulent Claims – A Machine Learning Approach. In Risk Insights: Vol. No. 1–2020. https://www.genre.com/content/dam/generalreinsuranceprogram/documents/ri20-1-en.pdf
• Sunita, M., Prasun, G., & Parita, S. (2018). Management of Fraud: Case of an Indian Insurance Company. Accounting and Finance Research, Sciedu Press, vol. 7(3), 1-18. https://ideas.repec.org/a/jfr/afr111/v7y2018i3p18.html