Table 6 Comparison of the adopted ML models for different CS ranges.

From: Machine learning and interactive GUI for concrete compressive strength prediction

CS range

Performance indices

MLR

MNLR

SVR

GEP

ANN

ANFIS

RF

AdaBoost

XGBoost

CatBoost

Low CS

R2

0.248

0.173

0.271

0.529

0.731

0.411

0.827

0.519

0.842

0.854

RMSE (MPa)

10.05

11.66

9.98

6.10

3.26

6.16

2.73

8.46

2.53

2.37

MAPE

0.614

0.778

0.609

0.364

0.111

0.339

0.111

0.615

0.109

0.098

Moderate CS

R2

0.211

0.204

0.205

0.450

0.862

0.596

0.925

0.571

0.922

0.932

RMSE (MPa)

9.65

8.97

10.16

6.61

3.19

5.61

2.19

5.48

2.24

2.08

MAPE

0.197

0.178

0.195

0.134

0.042

0.117

0.040

0.113

0.043

0.041

High CS

R2

0.125

0.001

0.093

0.249

0.744

0.505

0.801

0.414

0.886

0.925

RMSE (MPa)

13.28

15.91

13.32

10.58

4.40

8.11

3.99

8.46

2.95

2.32

MAPE

0.168

0.203

0.167

0.137

0.035

0.103

0.039

0.113

0.029

0.026

  1. *The bold values indicated the best predictive models across the ranges of CS.