AUTO-INSURANCE FRAUD DETECTION USING MACHINE LEARNING

ABSTRACT

This project explores the application of machine learning (ML) in auto-insurance fraud detection, addressing the rising challenge of fraudulent claims in the insurance industry. By leveraging advanced data preprocessing techniques, feature selection methods, and classification algorithms, the study develops a predictive model to distinguish between fraudulent and genuine claims. Various machine learning models, including Naive Bayes, Decision Trees, Support Vector Machines (SVM), and K-Nearest Neighbors (KNN), are implemented and evaluated based on accuracy, precision, recall, F1-score, and ROC-AUC curves. The results indicate that SVM and KNN achieved the highest accuracy of 94%, making them the most effective models for fraud detection. Despite the promising results, challenges such as data imbalance and computational complexity are encountered. The study concludes that machine learning can significantly enhance fraud detection efficiency, reducing financial losses for insurance companies.