One of the most significant annual expenses that a person has is their health insurance coverage. Health insurance accounts for one-third of GDP, and everyone needs medical treatment to varying degrees. Changes in med...One of the most significant annual expenses that a person has is their health insurance coverage. Health insurance accounts for one-third of GDP, and everyone needs medical treatment to varying degrees. Changes in medicine, pharmaceutical trends, and political factors are only a few of the many factors that cause annual fluctuations in healthcare costs. This paper describes how a system may analyse a person’s medical history to display their insurance plans and make predictions about their health insurance premiums. The performance of four ML models—XGBoost, Lasso, KNN, and Ridge—is evaluated using R2-score and RMSE. The analysis of medical health insurance cost prediction using Lasso regression, Ridge regression, and K-Nearest Neighbours (KNN), and XGBoost (XGB) highlights notable differences in performance. KNN has the lowest R2-score of 55.21 and an RMSE of 4431.1, indicating limited predictive ability. Ridge Regression improves on this by an R2-score of 78.38 but has a higher RMSE of 4652.06. Lasso Regression slightly edges out Ridge with an R2-score of 79.78, yet it suffers from an advanced RMSE of 5671.6. In contrast, XGBoost excels with the highest R2-score of 86.81 and the lowermost RMSE of 4450.4, demonstrating superior predictive accuracy and making it the most effective model for this task. The best method for accurately predicting health insurance premiums was XGBoost Regression. The findings beneficial for policymakers, insurers, and healthcare providers as they can use this information to allocate resources more efficiently and enhance cost-effectiveness in the healthcare industry.展开更多
文摘One of the most significant annual expenses that a person has is their health insurance coverage. Health insurance accounts for one-third of GDP, and everyone needs medical treatment to varying degrees. Changes in medicine, pharmaceutical trends, and political factors are only a few of the many factors that cause annual fluctuations in healthcare costs. This paper describes how a system may analyse a person’s medical history to display their insurance plans and make predictions about their health insurance premiums. The performance of four ML models—XGBoost, Lasso, KNN, and Ridge—is evaluated using R2-score and RMSE. The analysis of medical health insurance cost prediction using Lasso regression, Ridge regression, and K-Nearest Neighbours (KNN), and XGBoost (XGB) highlights notable differences in performance. KNN has the lowest R2-score of 55.21 and an RMSE of 4431.1, indicating limited predictive ability. Ridge Regression improves on this by an R2-score of 78.38 but has a higher RMSE of 4652.06. Lasso Regression slightly edges out Ridge with an R2-score of 79.78, yet it suffers from an advanced RMSE of 5671.6. In contrast, XGBoost excels with the highest R2-score of 86.81 and the lowermost RMSE of 4450.4, demonstrating superior predictive accuracy and making it the most effective model for this task. The best method for accurately predicting health insurance premiums was XGBoost Regression. The findings beneficial for policymakers, insurers, and healthcare providers as they can use this information to allocate resources more efficiently and enhance cost-effectiveness in the healthcare industry.
