Hybridized Extreme Learning Machine Model with Salp Swarm Algorithm: A Novel Predictive Model for Hydrological Application

Table 2

Performance evaluation for ELM and SSA-ELM, SVR, GRNN, and RF models.

Input combinations	Models	r	WI	RMSE (m^3·s^−1)	MAE (m^3 · s^−1)
M1	SSA-ELM	0.812	0.861	80.592	66.203
	ELM	0.81	0.82	98.272	87.012
	SVR	0.761	0.802	106.749	90.905
	GRNN	0.658	0.689	127.818	113.698
	RF	0.372	0.531	189.778	131.966
M2	SSA-ELM	0.809	0.850	80.516	62.185
	ELM	0.817	0.834	95.571	83.899
	SVR	0.715	0.769	111.228	95.14
	GRNN	0.468	0.54	125.187	103.551
	RF	0.373	0.549	175.765	142.347
M3	SSA-ELM	0.784	0.812	86.868	62.185
	ELM	0.815	0.814	104.307	92.544
	SVR	0.741	0.776	118.02	101.85
	GRNN	0.496	0.54	125.187	103.551
	RF	0.037	0.33	280.85	225.27
M4	SSA-ELM	0.800	0.855	87.84	69.069
	ELM	0.803	0.826	98.432	85.117
	SVR	0.728	0.769	119.539	106.585
	GRNN	0.346	0.507	133.522	114.063
	RF	0.127	0.39	1246.48	196.745
M5	SSA-ELM	0.801	0.856	87.659	67.886
	ELM	0.799	0.853	87.906	71.544
	SVR	0.715	0.768	124.155	108.367
	GRNN	0.248	0.416	135.35	112.606
	RF	0.222	0.0001	668.76	657.031