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

Table 3

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

Model	MAE	MAPE	COR
LSTM	13.01	97.32	0.31
GRU	16.97	127.54	0.40
SEQ2SEQ	14.83	97.16	0.41
Attention	14.68	105.19	0.32
A&A learning	11.21	80.31	0.55