Table 9 Comparative analysis with prior studies.

From: Estimating the surface chloride concentration of marine concrete utilizing advanced hybrid machine learning models

Reference

Models

Key features

R2

65

Ensemble

 

0.83

RF

 

0.81

Multi-layer perceptron ANN

 

0.76

SVR

 

0.47

Gaussian process regression

 

0.47

Linear regression

 

0.46

Gradient boosting

 

0.81

K-nearest neighbors

 

0.68

27

 

SHAP

 

Ensemble

 

0.95

Multi-layer perceptron

 

0.46

XGBoost

 

0.78

Gradient Boosting

 

0.80

RF

 

0.80

DT

 

0.76

SVR

 

0.71

Linear regression

 

0.36

66

CG-BP ANN

 

0.91

Present study

 

SHAP and PDP

 

Python GUI for immediate predictions

SVR-GWO

 

0.96

SVR-GTO

 

0.95

ANN-GWO

 

0.92

ANN-GTO

 

0.90