Author(s)

Yi Guan and Lei Tan

Corresponding Author

Yi Guan

Abstract

In recent years, the severe fog and haze weather in many areas of our country has made a tremendous impact on people's lives. According to the air quality and meteorological data of Beijing in recent years on the Internet, the occurrence of haze is predicted, and the role of various factors in haze prediction is analyzed. A variety of classification models including BP neural network are established, and the model is trained by cross-validation and the prediction results are obtained. Select different attribute groups for classification, combined with ROC curve, accuracy and other evaluation criteria, analyze the impact of different attribute groups on haze weather, and get the relationship between heating, transportation and haze weather. This work can provide theoretical support for haze prevention and control.

Keywords

Haze pollution; Data mining methed; Big data