Research on air quality prediction based on correlation analysis and XGBoost

As one of the primary concerns of people, air quality, its pollution and management is an important topic. Air pollution can cause harm to human health, ecological environment, and social economy, and its pollution level is affected by many factors. Exploring the factors of air pollutant concentration and predicting PM2.5 concentration and AQI index more accurately is a common concern for the scientific community and policymakers. In order to make a more accurate prediction of air quality, this paper addresses the problem of air quality prediction. The quantitative features are extracted based on the Pearson correlation model and stepwise regression model, and then the qualitative features are extracted based on the chi-square test. After the feature extraction, an XGBoost model is constructed and the extracted features are used as input to predict the future air quality. The obtained prediction accuracy is much higher than that of the support vector machine and BP neural network. The predicted R ² is 0.905.