Table 9 Summary of statistical evaluations for GEP, ANN and MEP.

From: Development of machine learning models for forecasting the strength of resilient modulus of subgrade soil: genetic and artificial neural network approaches

Parameters

MAE

R

RMSE

ρ

R2

MSE

U95

IA

SI

A10-index

VAF

Rank

GEP-Training

1.715

0.996

2.408

0.036

0.992

5.798

1.173

0.833

0.070

0.778

99.235

9.200

GEP-Testing

2.570

0.990

3.636

0.054

0.981

13.221

1.132

0.621

0.163

0.280

98.090

8.000

MEP-Training

2.499

0.990

3.821

0.057

0.980

14.601

1.164

0.583

0.113

0.679

97.971

7.970

MEP-Testing

3.284

0.983

4.889

0.073

0.966

23.901

1.150

0.326

0.215

0.260

96.587

6.500

ANN-Training

3.102

0.983

4.937

0.073

0.966

24.371

1.172

0.307

0.145

0.634

96.607

6.600

ANN-Testing

5.481

0.973

7.384

0.112

0.946

54.521

1.160

0.471

0.248

0.180

94.543

4.500

Back to article page