Table 6 Comparison of R2 values for pollutant PM2.5 across locations.

From: Deep learning framework for hourly air pollutants forecasting using encoding cyclical features across multiple monitoring sites in Beijing

 

DNN-Unencoded

CNN-Unencoded

DNN-Encoded

CNN-Encoded

Aotizhongxin

0.9579

0.9579

0.956

0.956

Changping

0.9425

0.9427

0.9425

0.943

Dongsi

0.9518

0.9534

0.9541

0.9531

Guanyuan

0.955

0.9559

0.956

0.956

Huairou

0.9397

0.9401

0.9394

0.9408

Nongzhanguan

0.9614

0.9608

0.9617

0.9612

Shunyi

0.9517

0.9518

0.9526

0.9511

Tiantan

0.9586

0.9592

0.9602

0.9608

Wanliu

0.9565

0.957

0.9569

0.957

Wanshouxigong

0.9477

0.9475

0.948

0.9483

Back to article page