Skip to main content

Table 7 Mean and standard deviation (in parentheses) of the number of features, calculated over 50 hold out runs, where the base learners are ANNs with M = 10 units in the hidden layer.

From: Improving prediction accuracy of tumor classification by reusing genes discarded during gene selection

  

Breast

Colon

Leukemia

Ovarian

Average

GA-FS

input

15042.4(3.5)

917.3(4.5)

3456.3(2.5)

9837.4(4.3)

7313.4(3.7)

discarded

9438.6(1.5)

1082.6(2.8)

3672.6(2.9)

5316.6(3.7)

4877.6(2.7)

 

H-MTL

input

12620.2(5.6)

1042.3(4.5)

4082.1(3.6)

8847.3(2.1)

6648.0(4.0)

output

4206.7(5.9)

347.5(2.3)

1360.7(3.1)

2949.1(4.3)

2216.0(3.9)

 

discarded

7654.1(3.5)

610.2(2.7)

1686.3(5.2)

3357.6(2.1)

3327.1(3.4)

 

GA-MTL

input

15153.3(2.5)

1041.1(4.1)

3435.4(4.3)

10034.5(3.3)

7416.1(3.6)

output

9327.7(2.7)

959.0(2.5)

3693.6(3.8)

5119.5(3.4)

4775.0(3.1)

 

GA-MTL-IR

input

12541.5(3.6)

842.2(4.5)

4325.6(2.8)

7984.2(2.1)

6423.4(3.3)

output

4285.4(4.3)

547.6(4.6)

1117.1(2.0)

3812.2(3.8)

2440.6(3.7)

 

discarded

7654.1(3.5)

610.2(2.7)

1686.3(5.2)

3357.6(2.1)

3327.1(3.4)

 

e-GA-MTL

input

12700.3(4.1)

854.7(4.1)

4147.1(2.9)

10453.2(3.5)

7038.8(3.7)

output

4154.4(4.5)

486.4(4.4)

1272.7(2.7)

2454.4(3.5)

2092.0(3.8)

 

discarded

7645.2(5.3)

660.0(5.7)

1846.2(2.4)

2489.2(3.7)

3160.2(4.3)

Â