Skip to main content

Table 5 Testing accuracies under different procedures for colon cancer data

From: A new regularized least squares support vector regression for gene selection

Procedures

Classifier

No. of genes

Accuracy

SD

A: Proposed selection and criterion

RLS-SVR

    

δ j ≥ 1/q

+KFDA

4

0.9250

0.0083

 

+SVM

4

0.9067

0.0082

∑ δ j ≥ 80%

+KFDA

5

0.9200

0.0163

 

+SVM

5

0.9183

0.0157

q genes

+KFDA

10

0.9400

0.0186

 

+SVM

10

0.9300

0.0167

B: Other selection procedures

EB

+KFDA

9

0.9283

0.0076

 

+SVM

9

0.9200

0.0194

SGS1

+KFDA

10

0.925

0.0083

 

+SVM

10

0.9100

0.0153

SGS2

+KFDA

10

0.9283

0.0076

 

+SVM

10

0.9100

0.0186

C: Selection and classification together

 

IFFS

5

0.8806

0.0167

 

SVM-RFE

8

0.9032

n.a.

 

EB

9

0.919

n.a.