Temporal Attention with Domain-Specific Graph Regularization for PM2.5 Forecasting

摘要

PM2.5 is an essential indicator of the air quality, and accurate forecasting of PM2.5 can help improve people’s daily life. However, the PM2.5 concentration among different cities correlates with each other in terms of both space and time, and it is a non-trivial task for traditional physical/statistical methods to model such complex spatial-temporal correlations. Besides, various factors from environments can have an impact on PM2.5 forecasting, thus it is necessary to consider complex information and utilize its correlations with PM2.5 concentration. In this paper, we propose a temporal attention network with domain-specific graph regularization to improve the prediction of PM2.5 concentration. Firstly, we consider both vertical and horizontal transportation of PM2.5 among different cities to enhance the representation ability of graph neural networks. Secondly, we use temporal attention mechanism to capture correlations with past information. We compare our proposed methods with other state-of-the-art methods such as MLP, LSTM, GRU, GC-LSTM and PM2.5-GNN, and experimental results verify the effectiveness and efficiency of the proposed method.