Deep Extreme Learning Machine and Its Application in EEG Classification

Table 7

Performance comparison of DELM with ELM, KELM, and MLELM on the BCI competition II dataset IB.

Dataset	Algorithm	#	Training accuracy	Testing accuracy	Training time (s)	Testing time (s)
BCI competition II dataset IB	Basic ELM	Average	1.0000 ± 0	0.5172 ± 0.0395	3.4636 ± 0.1255	2.0602 ± 0.0670
		Best	1.0000	0.6056	—	—
	KELM	—	1.0000	0.5333	0.0738	0.1285
	ML-ELM	Average	0.6145 ± 0.0306	0.5219 ± 0.0284	10.4225 ± 0.1134	0.3970 ± 0.0182
		Best	0.6750	0.5833	—	—
	DELM	Average	0.7151 ± 0.0485	0.5211 ± 0.0266	8.9814 ± 0.2085	0.2603 ± 0.0231
		Best	0.8450	0.6056	—	—