Skip to main content

Table 4 Averaged BER and AUC of the four classifiers on the benchmark data sets

From: AVC: Selecting discriminative features on basis of AUC by maximizing variable complementarity

  

AVC

FAST

ARCO

Relief

FROC

BER

NB

0.184

0.242

0.194

0.209

0.269

 

SVM

0.148

0.166

0.147

0.174

0.201

 

1-NN

0.163

0.213

0.162

0.178

0.254

 

C4.5

0.28

0.352

0.287

0.296

0.299

AUC

NB

0.924

0.894

0.911

0.894

0.881

 

SVM

0.896

0.908

0.913

0.886

0.898

 

1-NN

0.873

0.855

0.874

0.861

0.822

 

C4.5

0.798

0.766

0.791

0.788

0.793

  1. Bold data in the table reflect the minimal averaged BER and AUC of four classifiers in the feature subset selected by five compared feature selection methods on the benchmark data sets
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us