Improving Classification Performance through an Advanced Ensemble Based Heterogeneous Extreme Learning Machines

Table 8

Description of -Statistic experiments. ( is number of samples, #Att. is number of attributes, #Class is number of classes, and #Ens. is number of ensembles).

Dataset		#Att.	#Class	#Ens.
Liver	345	6	2	700
Satellite	1200	10	7	1300
Feature subspace method
Dataset	Wave
	700
#Att.	10
#Class	3
#Ens.	500
The set of 10 features was divided into permutations subsets (first 500 permutations) of 4, 4, and 2.