Machine Learning-Based Vehicle Insurance Claim Fraud Detection and Risk Assessment
A unified machine learning framework designed to improve the reliability and consistency of fraud detection across vehicle insurance claims is presented and two ensemble models, namely, Random Forest and Extreme Gradient Boosting are created to classify fraudulent claims and to estimate risk.