Mortality Prediction through Machine Learning: Insights from XGBoost and Random Forest Model

This study explores the use of machine learning algorithms to predict deaths caused by different risk factors in various countries. Using a large dataset that covers many years and regions, we applied models like XGBoost and Random Forest to see how well they could predict death rates based on health and environmental factors. The XGBoost model was very accurate with an $\text{R}^{2}$ of $\mathbf{0. 9 9 5 6}$, showing it could explain most of the variability in death rates. The Random Forest model also performed well with an $\text{R}^{2}$ of 0.9926. These findings show that machine learning can be very useful in health analysis, helping to understand the risk factors that lead to higher death rates. The study discusses how these results can inform public health policies and guide future research.