Skip to main content

Table 4 Comparison results of RP and SVM with FIRF feature vectors on human dataset

From: Identification of self-interacting proteins by integrating random projection classifier and finite impulse response filter

Testing set

Acc (%)

Sen (%)

PE (%)

MCC (%)

SVM + PSSM+FIRF

 1

96.32

55.09

100.00

72.78

 2

95.94

53.92

100.00

71.85

 3

96.37

55.32

100.00

72.95

 4

96.78

56.25

100.00

73.73

 5

95.66

51.60

100.00

70.18

Average

96.21 ± 0.43

54.44 ± 1.79

100.00 ± 0.00

72.30 ± 1.36

RP + PSSM+FIRF

 1

98.10

76.84

100.00

86.77

 2

97.76

74.51

100.00

85.28

 3

97.70

71.63

100.00

83.59

 4

98.01

73.05

100.00

84.57

 5

97.87

76.28

100.00

86.34

Average

97.89 ± 0.17

74.46 ± 2.18

100.00 ± 0.00

85.31 ± 1.29