Table 8 Performance analysis of the proposed model over SOTA methods of dataset 2.

From: A secure and efficient deep learning-based intrusion detection framework for the internet of vehicles

Metrics

CAN dataset

Proposed

DFSENet [15]

GAN [18]

TCAN-IDS [22]

XAIEM [23]

Sensitivity

0.973

0.945

0.96

0.948

0.933

Specificity

0.981

0.975

0.967

0.96

0.95

Accuracy

0.984

0.974

0.967

0.958

0.95

Precision

0.973

0.94

0.96

0.955

0.965

Recall

0.973

0.952

0.963

0.952

0.94

F-measure

0.972

0.96

0.95

0.948

0.935

NPV

0.982

0.96

0.963

0.957

0.95

FPR

0.018

0.025

0.035

0.031

0.04

FNR

0.092

0.085

0.07

0.08

0.095

MCC

0.972

0.96

0.965

0.95

0.94

Training Time (s)

550

600

650

620

640

Testing Time (s)

110

120

130

115

125

Inference Time (ms)

15

18

20

17

19

Back to article page