Skip to main content

Table 5 Comparison of performance for the SRBCT and Leukemia data sets using six classifiers with the same number of genes chosen by three gene selection methods

From: Discovery of dominant and dormant genes from expression data using a novel generalization of SNR for multi-class problems

Data Sets

Gene Selection Methods

Classifiers

m-fold of genes

   

1-fold

2-fold

3-fold

4-fold

5-fold

SRBCT

GDI.Dominant

OVO.SVM-L

10.1 ± 0.6

3.8 ± 0.4

1.5 ± 0.3

1.1 ± 0.2

1.0 ± 0.2

  

OVO.SVM-R

10.4 ± 0.6

3.5 ± 0.4

2.5 ± 0.3

2.6 ± 0.3

2.7 ± 0.4

  

OVA.SVM-L

9.0 ± 0.6

2.8 ± 0.3

1.0 ± 0.2

0.6 ± 0.2

0.6 ± 0.2

  

OVA.SVM-R

9.5 ± 0.7

2.5 ± 0.3

1.2 ± 0.3

1.5 ± 0.3

2.4 ± 0.4

  

NMC

8.2 ± 0.5

3.3 ± 0.4

1.0 ± 0.2

0.8 ± 0.2

0.5 ± 0.2

  

NNC

10.6 ± 0.6

2.9 ± 0.3

1.1 ± 0.2

0.9 ± 0.2

1.0 ± 0.2

 

OVA.SNR [12]

OVO.SVM-L

11.1 ± 0.7

3.8 ± 0.4

1.3 ± 0.3

0.9 ± 0.3

0.8 ± 0.3

  

OVO.SVM-R

11.8 ± 0.8

4.0 ± 0.5

3.2 ± 0.4

3.8 ± 0.5

3.5 ± 0.5

  

OVA.SVM-L

9.5 ± 0.7

3.2 ± 0.4

1.1 ± 0.3

0.7 ± 0.2

0.6 ± 0.2

  

OVA.SVM-R

10.7 ± 0.7

3.6 ± 0.5

1.9 ± 0.4

2.6 ± 0.4

3.0 ± 0.4

  

NMC

9.2 ± 0.6

3.9 ± 0.4

1.1 ± 0.2

0.9 ± 0.2

0.5 ± 0.2

  

NNC

10.4 ± 0.6

3.4 ± 0.4

1.2 ± 0.3

0.9 ± 0.2

0.8 ± 0.2

 

ANOVA+Correlation [11]

OVO.SVM-L

10.7 ± 0.6

3.4 ± 0.4

1.4 ± 0.3

0.6 ± 0.2

0.8 ± 0.2

  

OVO.SVM-R

10.8 ± 0.6

3.6 ± 0.4

2.1 ± 0.4

1.9 ± 0.4

1.8 ± 0.4

  

OVA.SVM-L

9.4 ± 0.6

2.8 ± 0.4

1.2 ± 0.3

0.4 ± 0.2

0.5 ± 0.2

  

OVA.SVM-R

8.8 ± 0.6

2.9 ± 0.4

1.6 ± 0.3

1.5 ± 0.3

1.4 ± 0.3

  

NMC

8.1 ± 0.6

3.1 ± 0.4

0.9 ± 0.2

0.5 ± 0.2

0.6 ± 0.2

  

NNC

10.0 ± 0.5

3.0 ± 0.4

1.0 ± 0.2

0.5 ± 0.2

0.7 ± 0.2

Leukemia

GDI.Dominant

OVO.SVM-L

13.4 ± 0.8

9.0 ± 0.5

7.3 ± 0.5

6.3 ± 0.5

5.8 ± 0.5

  

OVO.SVM-R

15.0 ± 0.9

9.0 ± 0.6

6.9 ± 0.4

6.1 ± 0.4

5.8 ± 0.5

  

OVA.SVM-L

13.8 ± 0.8

9.8 ± 0.6

8.2 ± 0.4

7.1 ± 0.5

6.4 ± 0.5

  

OVA.SVM-R

15.0 ± 0.9

9.2 ± 0.6

6.5 ± 0.5

6.0 ± 0.4

5.3 ± 0.4

  

NMC

13.8 ± 0.8

9.1 ± 0.6

7.1 ± 0.5

6.4 ± 0.5

5.9 ± 0.4

  

NNC

13.7 ± 0.8

9.6 ± 0.6

7.6 ± 0.5

7.0 ± 0.5

6.0 ± 0.5

 

OVA.SNR [12]

OVO.SVM-L

13.4 ± 0.8

9.1 ± 0.6

7.5 ± 0.5

7.0 ± 0.5

6.4 ± 0.4

  

OVO.SVM-R

13.5 ± 0.8

8.8 ± 0.7

7.2 ± 0.5

6.6 ± 0.4

6.4 ± 0.5

  

OVA.SVM-L

14.2 ± 0.8

10.9 ± 0.7

8.4 ± 0.5

7.2 ± 0.5

6.8 ± 0.4

  

OVA.SVM-R

13.6 ± 0.8

9.5 ± 0.6

7.2 ± 0.5

6.4 ± 0.4

6.1 ± 0.5

  

NMC

15.0 ± 0.7

8.8 ± 0.5

7.0 ± 0.5

6.7 ± 0.5

6.4 ± 0.5

  

NNC

13.5 ± 0.7

8.7 ± 0.5

7.8 ± 0.5

7.7 ± 0.5

7.2 ± 0.5

 

ANOVA+Correlation [11]

OVO.SVM-L

12.9 ± 0.8

10.2 ± 0.6

8.1 ± 0.5

7.9 ± 0.5

7.1 ± 0.5

  

OVO.SVM-R

13.5 ± 0.8

10.0 ± 0.7

7.7 ± 0.5

7.1 ± 0.5

6.9 ± 0.5

  

OVA.SVM-L

12.7 ± 0.8

11.6 ± 0.6

9.3 ± 0.5

8.5 ± 0.5

7.6 ± 0.5

  

OVA.SVM-R

12.7 ± 0.8

9.8 ± 0.6

8.0 ± 0.6

6.8 ± 0.5

6.4 ± 0.5

  

NMC

14.5 ± 0.8

9.4 ± 0.6

7.8 ± 0.6

7.1 ± 0.5

6.6 ± 0.5

  

NNC

12.3 ± 0.7

9.8 ± 0.6

8.8 ± 0.6

7.4 ± 0.5

6.9 ± 0.5

  1. Here m-fold corresponds to the case when m top most dominant genes are used for each class. For example, the column labeled 3-fold represents the results using 12 genes (3 dominant genes from each of the 4 classes) for the SRBCT data set. Similarly, for the Leukemia data set the same column represents results using 9 dominant genes as there are 3 classes.