Table 7 Comparative outcomes of the SAMFL-SCDCOA model with existing approaches under the UNSW-NB15 dataset.

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

Method

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

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

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

\({F}_{score}\)

KNN algorithm

92.79

94.92

83.07

88.72

MLP model

92.39

91.89

87.08

85.10

CNN classifier

94.23

92.35

85.13

85.15

SVM model

92.77

86.14

80.92

86.43

LSTM method

95.73

85.38

84.65

88.24

DE-VIT model

97.48

88.56

81.79

86.40

DBN algorithm

92.44

88.45

80.56

89.61

HMO-EVO

92.46

91.95

87.14

85.16

MBiLSTM-GRU

94.31

92.41

85.19

85.20

MRS-PFIDS

92.84

86.19

80.99

86.49

SAMFL-SCDCOA

99.04

95.13

89.77

91.69

Back to article page