Table 11 RMSE and MAE of the prediction approach in trend prediction.

From: A hybrid deep learning air pollution prediction approach based on neighborhood selection and spatio-temporal attention

Approach

RMSE

MAE

1–12 h

13–24 h

25–36 h

37–48 h

1–12 h

13–24 h

25–36 h

37–48 h

CNN

18.971

21.133

22.166

23.292

11.984

13.512

13.995

14.522

LSTM

18.069

20.110

21.398

22.748

11.127

12.921

13.669

14.529

CNN-LSTM

15.496

16.657

17.220

19.456

9.566

10.556

11.707

13.288

ConvLSTM

15.232

16.818

17.105

19.006

9.510

10.819

11.353

12.375

STARes-SaLSTM

14.729

15.326

16.350

16.884

9.728

9.935

10.027

10.587

KSC-ConvLSTM

10.513

10.077

10.492

10.844

6.919

6.376

6.700

6.939

Back to article page