Table 8 Performance comparison between the proposed model and existing methods on the CICIDS2017 dataset.

From: A hybrid intrusion detection model based on dynamic spatial-temporal graph neural network in in-vehicle networks

Method

Accuracy (%)

Precision (%)

Recall (%)

F1-score (%)

Training time (s)

Testing time (ms)

XGBoost

99.90

99.90

99.90

99.90

1620

0.2

IoVST

99.90

99.82

99.86

99.82

–

0.72

HNN

99.87

–

–

99.90

–

–

CNN

99.82

99.82

99.83

99.84

2658

1.5

SVM-B

98.92

98.93

99.46

–

–

–

GCN-2-Former

99.98

99.97

99.97

99.98

284

0.09

Back to article page