Table 7 Performance comparison between the proposed model and existing methods on the car hacking 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)

DeepSecDrive

98.13

98.59

98.02

98.30

–

0.26

DGIDS

99.63

99.92

99.60

99.74

–

117

Rec-CNN

99.90

99.90

–

–

5400

0.8

CNN

99.94

99.92

99.94

99.93

1680

5.96

SIDILDNG

95.84

97.73

–

96.61

–

0.007

GCN-2-Former

1.00

1.00

1.00

1.00

2143

0.09

Back to article page