Machine Learning-Based Regression Model to Predict Health Insurance Claim

In today’s world, where there are a lot of accidents and disasters happening, there is a need for a huge amount of money for health care as the cost of treatments and medicines requirements are very high. There are many cases where people can’t afford the treatment and lose their life. To prevent this, government along with many private banks tied up with numerous hospitals to set up many insurance agencies. People deposit some amount of money in the bank on a monthly or yearly basis in return for which the insurance agency provides money as health insurance. Costs of health insurance throughout the globe have increased critically in recent times. The increased trend results in the requirement of quick decision-making while providing better accuracy. This paper deals with the prediction regarding claiming health insurance i.e., whether a person claims his/her health insurance based on different factors. Using human brains for prediction agencies faces different errors which often results in extra administrative effort for reworking and finally leads to very little accuracy. Machine Learning algorithms can be the alternatives for the above-said problems as they can handle huge amounts of data. In this paper, Artificial Neural Networks (ANN), Decision Trees (DT), Support Vector Machines (SVM), and Logistic Regression Models (LRM) are applied to the dataset. The DT shows the best accuracy among all the applied algorithms.