Longer Time Span Air Pollution Prediction: The Attention and Autoencoder Hybrid Learning Model

Table 1

Experiment on the dataset of Shanghai with time span of 48, node size of 64, and data segmentation of [24, 72].

Model	MAE	MAPE	COR
LSTM	15.56	97.32	0.27
GRU	15.38	103.51	0.21
SEQ2SEQ	13.21	83.27	0.19
Attention	14.53	89.41	0.21
A&A learning	8.97	56.95	0.63