New techniques are required to effectively manage vehicle insurance claims due to the rising trend in claim severity and volume. One approach to resolving this issue is machine learning (ML).Car insurance firms have started implementing Machine Learning to boost the efficient interpretation of client data, greater knowledge of client demands to enhance customer service. With the help of Machine Learning algorithm, one can predict the car insurance claim. This is done in an effort to better serve their customers. This study examines how auto insurance firms use machine learning within their organizations and investigates how ML models may be used with large-scale insurance data. To forecast the recurrence of a claim, a variety of ML techniques is used, including Naive Bayes, KNN, Decision Tree, Random Forest, logistic regression, and XGBoost. The performances of these models are also assessed and contrasted. The outcomes demonstrated that Random Forest had superior accuracy than other approaches.