Table 7 Performance analysis of proposed model over SOTA methods of dataset 1.

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

Metrics

CIC-IDS 2017 dataset

Proposed

DFSENet [15]

GAN [18]

TCAN-IDS [22]

XAIEM [23]

Sensitivity

0.985

0.96

0.975

0.95

0.93

Specificity

0.99

0.965

0.97

0.963

0.95

Accuracy

0.991

0.98

0.975

0.965

0.955

Precision

0.985

0.96

0.955

0.945

0.94

Recall

0.985

0.95

0.963

0.94

0.93

F-measure

0.985

0.965

0.965

0.952

0.94

NPV

0.991

0.98

0.975

0.97

0.96

FPR

0.011

0.022

0.03

0.035

0.04

FNR

0.037

0.05

0.06

0.065

0.07

MCC

0.985

0.97

0.955

0.95

0.94

Training Time (s)

500

650

700

600

650

Testing Time (s)

100

120

140

110

130

Inference Time (ms)

14

18

22

20

19

Back to article page