Prediction of air pollution caused mortality exploiting machine learning models

Abstract

Air pollution is one of the most serious problems facing all mankind. It can not only cause various diseases and even lead to premature death of human beings. Especially, it could cause more harm to children, the elderly, and people who have been ill. Air pollution mainly occurs in developing countries with low income. Therefore, in order to better deal with the harm caused by air pollution, people need to predict the air pollution deaths. Based on the data set of air pollution deaths from 1990 to 2017, this paper investigated the performance of four machine learning models in predicting air pollution deaths. After the evaluation and comparison of four indicators, linear regression model and Ridge regression model are selected for further prediction considering their superior performance. This study finds that the overall trend of air pollution deaths worldwide is declining and will fall to zero by 2059 without other interference.