Soft computing techniques to predict the compressive strength of groundnut shell ash-blended concrete

Journal of Engineering and Applied Science

Table 3 Performance indicators for various ML models

	Train					Test					OBJ	Ranking
	R²	RMSE	MAE	SI	a20	R²	RMSE	MAE	SI	a20	OBJ	Ranking
LR	0.6083	5.11	3.82	0.30	0.55	0.6435	4.29	3.44	0.27	0.52	5.27	7
FQ	0.8654	3.00	2.34	0.18	0.69	0.7659	3.48	2.82	0.22	0.61	3.10	4
ANN	0.8697	2.95	2.34	0.17	0.69	0.7665	3.47	2.78	0.22	0.60	3.07	3
RFR	0.9079	2.48	1.80	0.15	0.81	0.8866	2.42	1.83	0.15	0.80	2.25	1
BDT	0.8849	2.77	2.12	0.16	0.72	0.8357	2.91	2.33	0.19	0.70	2.68	2
KNN	0.7490	4.09	2.80	0.24	0.64	0.7939	3.26	2.58	0.21	0.64	3.71	6
SVR	0.8020	3.63	2.32	0.21	0.73	0.8197	3.05	2.29	0.20	0.70	3.18	5