Deep Extreme Learning Machine and Its Application in EEG Classification

Table 4

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

Dataset	Algorithm	#	Training accuracy	Testing accuracy	Training time (s)	Testing time (s)
BCI competition II dataset IA	Basic ELM	Average	1.0000 ± 0	0.8609 ± 0.0187	3.3670 ± 0.0866	2.2361 ± 0.0498
		Best	1.0000	0.9078	—	—
	KELM	—	0.8582	0.9010	0.0754	0.1430
	ML-ELM	Average	0.7849 ± 0.0213	0.8642 ± 0.0216	9.2012 ± 0.1444	0.4820 ± 0.0272
		Best	0.8358	0.9113	—	—
	DELM	Average	0.7515 ± 0.0161	0.8650 ± 0.0224	6.7438 ± 0.2099	0.2932 ± 0.0290
		Best	0.7873	0.9181	—	—