<![CDATA[Insurance is an effective way to guard against potential loss. Risk management is primarily employed to protect against the risk of a financial loss. Risk and uncertainty are inevitable parts of life, and the pace of life has led to a rise in these risks and uncertainties. Health insurance pricing has emerged as one of the essential fields of this study following the coronavirus pandemic. The anticipated outcomes from this study will be applied to guarantee that an insurance company's goal for its health insurance packages is within the range of profitability so that the insurance company will also choose the most price-effective course of action. The US Health Insurance dataset was utilized for this study. This health insurance pricing prediction aims to examine four different types of regression-based machine learning algorithms: multiple linear regression, ridge regression, XGBoost regression, and random forest regression. The implemented model's performance is assessed using four evaluation metrics: MAE, MSE, RMSE, and R2 score. Random forest regression outperforms all other algorithms in terms of all four evaluation metrics. The best machine learning algorithm, random forest, is further enhanced with hyperparameter tuning. Random forest with hyperparameter tuning performs better for three evaluation metrics except for MAE. To gain further insights, data visualizations are also implemented to showcase the importance of features and the differences between actual and predicted prices for all the data points.]]>
Copyrights © 2024
