• Version
• Download 4
• File Size 321.74 KB
• File Count 1
• Create Date 24 March 2026
• Last Updated 24 March 2026

Using Policyholder and Incident Risk Factors in Predictive Modelling of Auto Insurance Claims

Dr. K. Satyam1, Veera Leela Manikanta2
1Associate Professor, Department of MCA, Annamacharya Institute of Technology & Sciences, Tirupati,Andhra Pradesh, India.
2 Post Graduate, Department of MCA, Annamacharya Institute of Technology & Sciences, Tirupati, AndhraPradesh, India.

ABSTRACT:Effective claim analysis is a major difficulty for insurance firms due to the substantial rise in auto insurance policies and claim requests brought about by the automobile industry's rapid growth. To reduce financial losses and increase operational effectiveness, accurate claim severity prediction and fraudulent claim detection are crucial. Using policyholder data, vehicle attributes, and incident-related variables, this study suggests a machine learning-based method for modelling auto insurance claims. Age, insurance type, premium amount, car information, accident severity, and claim costs are just a few of the many attributes included in the dataset. To improve the quality of the data, preprocessing methods such as addressing missing values, encoding categorical variables, and feature normalisation were used. To examine trends in the dataset and forecast claim outcomes, including fraud detection, a number of classification algorithms were put into practice. The experimental findings show that, in comparison to conventional techniques, machine learning models may successfully detect high-risk claims and increaseforecast accuracy. Insurance businesses can automate decision-making, identify fraudulent activity, and streamline claim management procedures with the help of the suggested system.
Keywords:Auto Insurance, Machine Learning, Claim Prediction, Fraud Detection, Risk Analysis, Classification Algorithms, Data Preprocessing, Predictive Modeling