Skip to main content

Table 2 The highest recognition accuracies on gene expression databases when selecting features within the top 100 ranked features obtained by three different classifiers

From: Ranked selection of nearest discriminating features

  

Nearest neighbor

SVM

Naive Bayes

Database

Total number of features

Accuracy (%)

Selected number of features

Accuracy (%)

Selected number of features

Accuracy (%)

Selected number of features

GLI-85

22283

88.3±2.9

3

86.5±5.2

2

89.1±2.9

3

GLA-BRA-180

4915

65.3±4.6

45

66.7±4.8

6

68.4±5.1

37

CLL-SUB-111

11340

74.9±2.6

1

65.6±5.5

78

66.5±8.3

50

TOX-171

5748

69.6±4.4

89

78.5±5.5

71

61.5±5.1

68

SMK-CAN-187

19993

71.2±1.7

1

73.2±3.2

48

70.8±4.0

52