Improving ELM-Based Service Quality Prediction by Concise Feature Extraction

Table 4

Accuracy comparison of different algorithms.

Algorithm	# Attributes	Training time	Testing time	Accuracy	-test
F-ELM	6	1.15	0.26	82.25%	N/A
ELM	12	1.82	0.52	81.2%	5.103
SVM	6	2.3	0.69	82.1%	5.211
SVM	12	2.9	0.89	80.55%	5.060
CART	6	3.1	0.67	74.1%	5.533
CART	12	3.5	0.88	72.1%	5.667
J48	6	3.41	0.96	66.72%	6.025
J48	12	4.01	1.2	63.77%	6.222
Treenet	6	2.22	0.74	77.2%	5.732
Treenet	12	2.79	0.99	75.4%	5.547
BPNN	6	1.87	0.77	63.1%	6.267
BPNN	12	2.12	0.97	60.1%	6.467