Skip to main content

Table 2 Performance comparison of classifiers under different percentage of pseudo-negative samples on the CMC data

From: How to balance the bioinformatics data: pseudo-negative sampling

Percentage

Classifier

Sen(%)

Spe(%)

Acc(%)

MCC

0

DA

9.38

97.81

77.8

0.156

 

AdaBoost

21.37

94.48

77.94

0.226

 

RF

28.19

92.8

78.2

0.27

 

NN

27.01

87.09

73.52

0.161

10

DA

17.6

94.85

75.7

0.198

 

AdaBoost

25.76

93.58

76.78

0.266

 

RF

39.22

91.77

78.75

0.369

 

NN

40.92

86.98

75.55

0.302

20

DA

37.35

91.71

76.94

0.351

 

AdaBoost

40.03

91.24

77.33

0.36

 

RF

43.94

91.24

78.41

0.404

 

NN

47.28

87.22

76.38

0.368

30

DA

52.46

88.34

77.8

0.438

 

AdaBoost

50.89

88.83

77.67

0.431

 

RF

50.87

89.98

78.48

0.448

 

NN

53.39

87.86

77.73

0.439

40

DA

59.46

87.21

78.43

0.485

 

AdaBoost

56.01

87.61

77.61

0.461

 

RF

56.45

90.27

79.57

0.505

 

NN

54.94

86.68

76.64

0.439

50

DA

66.78

85.42

79.08

0.530

 

AdaBoost

64.01

87.37

79.42

0.531

 

RF

62

88.71

79.63

0.532

 

NN

61.02

87.38

78.42

0.505