Lightweight Intrusion Detection Model of the Internet of Things with Hybrid Cloud-Fog Computing

Table 6

Experimental results of ConvNeXt, ConvNeXt-DenseNet, ConvNeXt-GhostNet, and the proposed model on ToN-IoT and BoT-IoT datasets.

Dataset	Model	Accuracy	Precision	Recall	F1-score	FAR	AUC
TON-IoT	ConvNeXt	0.9996	0.9997	0.9980	0.9989	0.0001	0.9998
	ConvNeXt-DenseNet	0.9996	0.9995	0.9984	0.9990	0.0001	0.9997
	ConvNeXt-GhostNet	0.9998	0.9996	0.9988	0.9992	0.0000	1.0000
	Proposed model	0.9998	0.9997	0.9987	0.9992	0.0000	0.9989
BoT-IoT	ConvNeXt	1.0000	0.9960	0.9923	0.9941	0.0000	1.0000
	ConvNeXt-DenseNet	1.0000	0.9961	0.9961	0.9961	0.0000	1.0000
	ConvNeXt-GhostNet	1.0000	1.0000	0.9739	0.9863	0.0000	0.9999
	Proposed model	1.0000	0.9962	1.0000	0.9981	0.0000	1.0000

The bold value in the table means that the corresponding model is optimal on this performance metric.