Table 3 Performance measures for model application on test data.

From: Modelling the urban acoustic environment using land use-based gradient boosting

 

SALVETest (n = 117)

Be-MoVe (n = 22)

 

MAE

RMSE

R2

nMAE

nRMSE

MAE

RMSE

R2

nMAE

nRMSE

LAeq

3.8

4.8

0.518

0.134

0.168

3.5

4.4

0.458

0.167

0.210

Articulation Index (RMS)

0.056

0.078

0.609

0.114

0.160

0.123

0.156

–0.149

0.249

0.316

BIO

0.652

0.851

0.300

0.122

0.159

1.102

1.242

–3.692

0.491

0.555

Link Density

0.140

0.172

0.273

0.158

0.194

0.156

0.205

0.312

0.184

0.242

Sharpness (Max)

0.216

0.290

0.256

0.132

0.178

0.205

0.239

–0.673

0.257

0.299

  1. For SALVETest the model trained on the SALVETrain model is used. For Be-MoVE, the SALVEAll model is used.
  2. MAE Mean Absolute Error, RMSE Root Mean Square Error, R2=Coefficient of Determination, nMAE MAE/Range, nRMSE RMSE/Range.
  3. The unit of LAeq is dB(A). The unit of Sharpness is Acum.
Back to article page