Predictive Modelling of Healthcare Insurance Costs Using Machine Learning

With more healthcare spending comes the added demand for predictive models, which would deliver forecasts of medical insurance spending as well as propose significant determining factors. Machine learning is used here to analyze the bills in healthcare insurance based on demographic and lifestyle factors such as age, BMI, smoking status, and geography. Based on the Medical Insurance Cost Prediction dataset, three regression models—Linear Regression, Random Forest Regression, and Gradient Boosting Regression—were employed to forecast insurance charges. Based on the outcome, the most significant variable influencing medical spending is revealed to be smoking status, followed by BMI and age. Among the models employed, Gradient Boosting Regression had the maximum predictive capability, outperforming Linear Regression, which struggled with complex relationships, and Random Forest Regression, which experienced some overfitting. The study highlights the ability of machine learning to enhance insurance pricing for optimization, enabling enhanced risk assessment by providers and decision-making by individuals. The research promotes the optimization of cost estimation methods in the healthcare sector based on data insights.