Prediction and lag analysis of public concern about air pollution based on gray relation analysis and bidirectional long short-term memory

AbstractWith the comprehensive green transformation of China ’s economic development, the people’s demand for a beautiful ecological environment is growing, and air pollution has become one of the focuses of public attention. The widespread use of the Internet has provided new ways for the public to understand environmental pollution. In this paper, a com bined gray relation analysis and bidirectional long short-term memory model is proposed for predicting public attention about air pollution in three cities: Beijing, Shanghai, and Guangzhou. The prediction of public attention is influenced by a variety of factors, and nine influencing factors were s elected for gray correlation analysis in this study. Subsequently, seven key factors with strong correlations were selected as input indicators for the prediction model to predict public attention. When compared with the results of other models, the combined GRA-BiLSTM model proposed in this study h as good predictive performance. Finally, taking into account the effect of lags, lagging in the input indicators can significantly improve the predictive performance of the model, verifying the existence of lags in public concern about air pollution. The environmental protection department can pay a ttention to the public attention about air pollution to better control it.Graphical Abstract
Source: Air Quality, Atmosphere and Health - Category: Environmental Health Source Type: research