Skip to main content

Table 5 Results of the SD datasets analysis

From: RGIFE: a ranked guided iterative feature elimination heuristic for the identification of biomarkers

Dataset

Metrics

RGIFE-Min

RGIFE-Max

RGIFE-Union

CFS

ReliefF

SVM-RFE

Chi-Square

L1

SD1

Selected

113.3

253.6

289.5

24.3

289.5

289.5

289.5

144.2

 

OPT(2)

0.2

0.8

0.9

1.5

2.0

2.0

1.7

2.0

 

Redundant

0.0

2.7

2.7

0.3

9.0

8.7

5.4

5.3

 

Irrelevant

114.1

248.4

284.2

23.1

270.5

271

278.5

132.6

SD2

Selected

103.4

279.7

319.4

23.1

319.4

319.4

319.4

137.1

 

OPT(4)

0.6

1.1

1.2

2.7

3.9

4.0

2.8

4.0

 

Redundant

0.6

2.4

2.6

0.1

9

8.9

3.6

3.5

 

Irrelevant

102.6

271.4

310.4

20.7

281.4

281.4

301.8

117.9

SD3

Selected

114.6

284.3

337.3

24.4

337.3

337.3

337.3

143.4

 

OPT(6)

1.0

2.6

3.8

3.4

4.8

4.2

3.5

6.0

 

Redundant

1.0

4.1

6.1

0.1

9.0

4.0

7.4

3.5

 

Irrelevant

113.2

267.8

312.9

21.1

292.0

306.6

309.2

119.2

  1. The values are averaged from a 10-fold cross-validation. OPT(x) indicates the average number of selected features within the optimal subset