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

Table 8

Experimental results of TON-IoT in various types.

Model	Metrics	0	1	2	3	4	5	6	7	8	9
TP2SF [17]	Precision	0.9600	0.9200	1.0000	0.9200	1.0000	0.9800	1.0000	0.9800	0.9900	0.9500
	Recall	1.0000	0.9200	0.9900	0.9100	0.8000	1.0000	0.9700	0.9900	0.9100	0.7500
	F1-score	0.9800	0.9200	0.9900	0.9200	0.8900	0.9900	0.9800	0.9800	0.9500	0.8400
Sp2f [18]	Precision	0.9991	0.9875	0.9976	0.9445	1.0000	1.0000	0.9983	0.9978	1.0000	0.9954
	Recall	0.9988	0.9429	0.9979	0.9870	0.9675	1.0000	0.9957	1.0000	1.0000	0.9970
	F1-score	0.9989	0.9647	0.9978	0.9653	0.9834	1.0000	0.9970	0.9989	1.0000	0.9970
ConvNeXt	Accuracy	1.0000	1.0000	1.0000	1.0000	1.0000	0.9997	0.9999	0.9998	1.0000	1.0000
	Precision	1.0000	0.9994	1.0000	1.0000	1.0000	0.9995	0.9989	0.9994	1.0000	1.0000
	Recall	1.0000	0.9994	1.0000	0.9989	0.9890	1.0000	0.9989	0.9948	1.0000	0.9994
	F1-score	1.0000	0.9994	1.0000	0.9994	0.9945	0.9997	0.9989	0.9971	1.0000	0.9997
	FAR	0.0000	0.0000	0.0000	0.0000	0.0000	0.0009	0.0001	0.0000	0.0000	0.0000
Proposed model	Accuracy	1.0000	1.0000	1.0000	0.9999	1.0000	0.9998	1.0000	1.0000	1.0000	1.0000
	Precision	1.0000	1.0000	1.0000	0.9994	1.0000	0.9999	0.9989	0.9994	0.9994	1.0000
	Recall	1.0000	0.9994	1.0000	0.9989	0.9890	0.9999	1.0000	1.0000	1.0000	1.0000
	F1-score	1.0000	0.9997	1.0000	0.9991	0.9945	0.9999	0.9994	0.9997	0.9997	1.0000
	FAR	0.0000	0.0000	0.0000	0.0000	0.0000	0.0002	0.0001	0.0000	0.0000	0.0000

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