Table 2 Performance evaluation for different algorithms on different test set sizes.

From: Inferring linear-B cell epitopes using 2-step metaheuristic variant-feature selection using genetic algorithm

Sr. no.

Test set size

 

Ensemble DL 4

iLBE 8

SVM 13

Proposed

1

Small set (2000–5000 epitopes)

ACC.

78.63

61.91

63.91

97.2

Precision

52.92

48.93

41.03

65.41

Recall

52.26

48.32

40.52

64.6

AUC

79.31

62.44

64.46

98.03

F-Measure

52.59

48.62

40.78

65.01

2

Medium set (6000–10,000 epitopes)

ACC.

80.01

62.99

65.03

98.9

Precision

53.85

49.79

41.75

66.56

Recall

53.17

49.17

41.23

65.73

AUC

80.7

63.54

65.59

99.75

F-Measure

53.51

49.47

41.49

66.14

3

Large set (11,000–13,000 epitopes)

ACC.

80.11

63.08

65.12

99.03

Precision

53.92

49.85

41.8

66.65

Recall

53.24

49.23

41.28

65.82

AUC

80.8

63.62

65.68

99.88

F-Measure

53.58

49.54

41.54

66.23

4

Very large set (14,000–16,000 epitopes)

ACC.

80.12

63.08

65.12

99.04

Precision

53.92

49.86

41.81

66.65

Recall

53.25

49.24

41.29

65.82

AUC

80.81

63.62

65.68

99.89

F-Measure

53.58

49.54

41.55

66.24

  1. #Bold value indicate the highest value.