Protein Sequence Classification with Improved Extreme Learning Machine Algorithms

Table 4

Training time comparisons of different classifiers: randomly generated training and testing datasets from the mixed protein sequences, where L, S, and G stand for the linear kernel, the sigmoid kernel, and the Gaussian kernel, respectively.

Methods	L		S		G
	Training time (s)	Speedup	Training time (s)	Speedup	Training time (s)	Speedup
VOP-ELM	5.2799	266.44	83.55	11.38	92.00	12.43
V-ELM	0.6326	2223.83	0.6917	1374.51	0.9282	1232.17
OP-ELM	1.8627	755.24	30.93	30.74	32.59	35.09
SVM	278.35	5.05	284.30	3.34	272.41	4.2
BP	1406.8	1	950.75	1	1143.70	1
ELM	0.0746	18858	0.0861	11042	0.0983	11635