Table 6 Comparative analysis of CDMFL-AIDCNN method under the UNSW-NB15 dataset25,26, and27.

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

UNSW-NB15 dataset

Model

\(\:\text{Accu}_{\text{y}}\)

\(\:\text{Prec}_{\text{n}}\)

\(\:\text{Reca}_{\text{l}}\)

\(\:{\text{F}}_{\text{measure}}\)

Decision Tree

89.52

80.03

83.11

88.94

Random Forest

90.98

81.29

79.57

78.51

DT-XGB Classifier

95.79

90.71

79.18

84.60

Random Forest-FS

88.73

85.33

90.54

82.53

Logistic Regression

89.99

77.56

81.74

86.50

KNN + XGBoost

97.31

91.14

78.00

89.66

SVM Method

95.29

84.00

92.34

91.94

CDMFL-AIDCNN

98.64

93.82

93.52

93.65

Back to article page