Determination of the relationship between respiratory diseases and pollutants in the atmospheric air of the city using machine learning methods
Please login to view abstract download link
Air pollution leads to ecosystem disruption and significant economic and social damage and affects the health of the population [1]. To identify the link between respiratory diseases and emissions of pollutants in the atmospheric air of Almaty, machine learning models were developed to identify the growth of primary morbidity, and priority pollutants affecting certain respiratory diseases are determined based on data from 2017 to 2022. Such methods as the random forest method, the multilayer perceptron learning method, the k-nearest neighbor method and the support vector method for constructing the highest quality model are considered [2, 3]. Metric estimates of these methods are considered. The models that showed the best result for 8 types of diseases were selected and the pollutants that have the greatest impact on them were identified. The study revealed a high correlation between morbidity and urban air pollution. Recommendations were proposed to reduce pollutants in the city's airspace.
