Table 3 Attack detection outcomes of the CLAFS-ODLCD method at 70%TRAPH and 30%TESPH.

From: Deep learning with leagues championship algorithm based intrusion detection on cybersecurity driven industrial IoT systems

Classes

\(Acc{u}_{y}\)

\(Sen{s}_{y}\)

\(Spe{c}_{y}\)

\({F}_{score}\)

\(AU{C}_{score}\)

Kappa

TRAPH (70%)

 Normal

98.23

99.28

87.92

99.03

93.60

93.67

 Blackhole

99.27

85.11

99.66

86.12

92.38

92.44

 Grayhole

99.28

89.63

99.67

90.68

94.65

94.71

 Flooding

99.49

62.75

99.83

69.09

81.29

81.35

 Scheduling Attacks

99.35

77.61

99.74

80.81

88.68

88.73

 Average

99.13

82.88

97.36

85.15

90.12

90.18

TESPH (30%)

 Normal

98.25

99.27

88.23

99.04

93.75

93.80

 Blackhole

99.23

84.10

99.66

85.64

91.88

91.95

 Grayhole

99.30

89.85

99.69

91.01

94.77

94.82

 Flooding

99.53

64.21

99.83

69.56

82.02

82.09

 Scheduling Attacks

99.33

78.46

99.71

80.77

89.08

89.13

 Average

99.13

83.18

97.42

85.21

90.30

90.36