Table 2 Cyberattack detection of SAMFL-SCDCOA technique on the CICIDS-2017 dataset.

From: Leveraging self attention driven gated recurrent unit with crocodile optimization algorithm for cyberattack detection using federated learning framework

Class labels

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

\(Pre{c}_{n}\)

\(Rec{a}_{l}\)

\({F}_{score}\)

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

TRPH (70%)

Benign

98.70

98.05

96.29

97.17

97.86

DDoS

99.13

97.67

98.42

98.04

98.87

DoS

99.18

97.84

98.57

98.20

98.96

Bot

99.26

98.40

97.48

97.94

98.57

Web attack

98.92

95.28

96.98

96.12

98.11

Average

99.04

97.45

97.55

97.49

98.47

TSPH (30%)

Benign

98.69

98.15

95.83

96.98

97.66

DDoS

99.03

97.62

98.36

97.99

98.80

DoS

99.42

98.81

98.68

98.74

99.16

Bot

99.51

98.61

98.61

98.61

99.16

Web attack

99.03

95.20

97.76

96.46

98.49

Average

99.14

97.68

97.85

97.76

98.66