Comparative Analysis of Gradient-Boosting Ensembles for Estimation of Compressive Strength of Quaternary Blend Concrete

Mustapha, Ismail B.; Abdulkareem, Muyideen; Jassam, Taha M.; AlAteah, Ali H.; Al-Sodani, Khaled A. Alawi; Al-Tholaia, Mohammed M. H.; Nabus, Hatem; Alih, Sophia C.; Abdulkareem, Zainab; Ganiyu, Abideen

doi:10.1186/s40069-023-00653-w

International Journal of Concrete Structures and Materials

Table 5 Comparison with previous studies

From: Comparative Analysis of Gradient-Boosting Ensembles for Estimation of Compressive Strength of Quaternary Blend Concrete

Researcher	Algorithm	Data set size	R²	RMSE	MAE	MAPE
		Based on average performance/multiple cross validation
(Chou et al., 2011)	ANN	1030	0.909	5.030		0.109
	SVM		0.886	5.619		0.128
	Multiple regression (MR)		0.611	10.429		0.317
	Multiple Additive Regression Tree (MART)		0.911	4.949		0.139
	Bagging Regression Tree (BRT)		0.890	5.572		0.142
(Erdal, 2013)	Decision Tree (DT)	1030	0.818
	Bagging DT		0.879
	Gradient-boosted DT		0.889
	Random Sub-spaced DT		0.868
(Chou et al., 2014)	SVM	1030		5.59	3.75	0.12
	Stacked CART + SVM + LR			5.08	3.52	0.12
(Feng et al., 2020)	Adaboost	1030	0.952	4.856	3.205	0.114
(Farooq et al., 2021)	Modified Random Forest Ensemble	1030	0.923	4.6	3.23
(Chen et al., 2021)	CNN	1030	0.97	3.98	2.68
	CNN–AP		0.97	4.09	2.92
	CNN–MP		0.96	4.18	2.89
This Study	XGB	1030	0.947	3.764	2.437	0.085
	LGBM		0.947	3.764	2.439	0.086
	CBT		0.951	3.605	2.246	0.077
	GBR		0.944	3.841	2.425	0.084
		Based on single cross validation
(Erdal et al., 2013)	ANN	1030	0.909	5.57	4.18
	Bagged ANN		0.928	4.87	3.60
	Gradient-boosted ANN		0.927	5.24	4.09
	Wavelength-bagged ANN		0.94	4.54	3.30
	Wavelength gradient-boosted ANN		0.953	5.75	4.83
(Silva et al., 2020)	ANN	1030	0.89	5.9
	SVM		0.83	7.5
	Random forest		0.90	5.6
(Dao et al., 2020a, 2020b)	Gaussian Process Regression (GPR-52)	1030	0.884	5.702	4.058
	Gaussian Process Regression (GPR-32)		0.888	5.597	3.913
	GPR using Exponential Kernel (GPR–EXP)		0.888	5.600	3.924
	GPR using Square Exponential Kernel (GPR–SQEXP)		0.878	5.849	4.242
	GPR using Rational Quadratic Kernel (GPR–RSQ)		0.880	5.793	4.182
	Levenberg–Marquardt ANN		0.890	5.447	4.274
(Feng et al., 2020)	Adaboost	1030	0.982	2.20	1.64	0.0678
(Salami et al., 2021)	LSSVM–CSA	1030	0.954	3.335
	GP		0.894	4.662
This Study	XGB	1030	0.976	2.497	1.903	0.074
	LGBM		0.974	2.596	2.007	0.079
	CBT		0.984	2.071	1.597	0.063
	GBR		0.973	2.664	1.901	0.072

Back to article page