Table 6.

Wilcoxon signed-rank test: proposed ensemble versus individual classifiers on Statlog dataset

	RF	LR	SVM	ELM	KNN
E
Z-value${}^1$	− 2.805	− 2.805	− 2.670	− 2.670	− 2.803
p value	$0.{002}^{\ast \ast }$	$0.{002}^{\ast \ast }$	$0.{004}^{\ast \ast }$	$0.{004}^{\ast \ast }$	$0.{002}^{\ast \ast }$
Precision
Z-value	− 2.692	− 2.829	− 2.680	− 2.677	− 2.807
p value	$0.{004}^{\ast \ast }$	$0.{002}^{\ast \ast }$	$0.{004}^{\ast \ast }$	$0.{004}^{\ast \ast }$	$0.{002}^{\ast \ast }$
Recall
Z-value	− 2.374	− 2.527	− 2.388	− 2.670	− 2.536
p value	$0.{016}^{\ast }$	$0.{008}^{\ast \ast }$	$0.{016}^{\ast }$	$0.{004}^{\ast \ast }$	$0.{008}^{\ast \ast }$
G-mean
Z-value	− 2.803	− 2.803	− 2.666	− 2.666	− 2.803
p value	$0.{002}^{\ast \ast }$	$0.{002}^{\ast \ast }$	$0.{004}^{\ast \ast }$	$0.{004}^{\ast \ast }$	$0.{002}^{\ast \ast }$
MC
Z-value	− 2.654	− 2.805	− 2.670	− 2.670	− 2.803
p value	$0.{006}^{\ast \ast }$	$0.{002}^{\ast \ast }$	$0.{004}^{\ast \ast }$	$0.{004}^{\ast \ast }$	$0.{002}^{\ast \ast }$
Specificity
Z-value	− 1.825	− 2.243	− 2.371	− 2.673	− 2.675
p value	0.086	$0.{023}^{\ast }$	$0.{016}^{\ast }$	$0.{004}^{\ast \ast }$	$0.{004}^{\ast \ast }$
AUC
Z-value	− 2.547	− 2.599	− 2.668	− 2.666	− 2.666
p value	$0.{008}^{\ast \ast }$	$0.{006}^{\ast \ast }$	$0.{004}^{\ast \ast }$	$0.{004}^{\ast \ast }$	$0.{004}^{\ast \ast }$

${}^1$The value of Wilcoxon statistics after standardization

* $p<0.05$** $p<0.01$