Leveraging Data Analytics to Strengthen Public Health and Global Economic Sustainability

Long-term economic viability and social development can be influenced by public health outcomes. Data analytic and machine learning methods are used to identify factors influencing public health due to the increasing availability of health-related data. This paper explores the key factors determining life expectancy using an open-source Life Expectancy dataset‚ to create an avenue for data driven public healthcare decisions. Exploratory data analysis is used to show the correlation between adult mortality‚ infant mortality‚ gross domestic product (GDP) per capita and schooling and their effect on life expectancy. Random Forest‚ Support Vector Regression (SVR)‚ and XGBoost are three different machine learning models that show that the proposed ensemble models are efficient in predicting life expectancy. The 3 models are XGBoost with R-squared of 0.998‚ Random Forest with R-squared of 0.997 and Support Vector Regression with R-squared of 0.980. The 3 most important features in prediction of life expectancy are under-five deaths‚ adult deaths and infant deaths. The study shows that predictive analytics can be a useful tool to assist policy makers to improve public health and sustainable economic growth.