Table 3 Comparative analysis of the CDMFL-AIDCNN method under the CIC-IDS-2017 dataset25,26, and27.

From: An efficient trustworthy cyberattack defence mechanism system for self guided federated learning framework using attention induced deep convolution neural networks

CIC-IDS-2017

Methods

Accuracy

F1-Score

Recall

Precision

MLP Algorithm

88.88

88.75

86.38

88.56

BBB-BAE-Homo

94.77

92.23

93.46

91.84

MCD-BAE-Hetero-Last

93.62

91.96

91.67

91.49

LSTM Classifier

91.02

91.09

89.96

92.56

1D-CNN Methodology

95.43

92.57

90.28

93.26

Deep-GFL Model

94.69

94.60

93.96

92.97

DBN Approach

95.59

92.58

92.88

88.87

ENIDS-IV Model

98.27

93.11

92.55

93.07

CDMFL-AIDCNN

99.07

94.65

94.56

94.75

Back to article page