Table 10 Comparative experimental results of FSLLM gramework.

Method	Dataset	Features	Binary cassification		Multi-classification
Method	Dataset	Features	F1-weighted	Accuracy	F1-weighted	Accuracy
Nguyen et al.²²	NF-CSE-CIC- IDS2018-v2	All	0.995	0.995	0.992	-
Nguyen et al.²²	NF-UNSW-NB15-v2	All	0.997	0.997	0.990	-
Li et al.³²	NF-ToN-IoT-v2	All	-	-	0.988	0.988
Li et al.³²	NF-UNSW-NB15-v2	All	-	-	0.890	0.874
Sarhan et al.⁸	NF-CSE-CIC- IDS2018-v2	All	0.889	0.891	-	-
Sarhan et al.⁸	NF-UNSW-NB15-v2	All	0.983	0.982	-	-
Wang et al.²³	NF-ToN-IoT-v2	All	0.979	0.966	0.914	-
Wang et al.²³	NF-BoT-IoT-v2	All	0.987	0.979	0.926	-
Termos et al.⁷	CIC-ToN-IoT	All	0.986	0.989	0.806	0.857
FSLLM	NF-CSE-CIC- IDS2018-v2	9	0.995	0.995	0.988	0.988
	NF-ToN-IoT-v2	7	0.990	0.990	0.958	0.958
	NF-UNSW-NB15-v2	9	0.997	0.997	0.992	0.992
	NF-BoT-IoT-v2	9	0.997	0.997	0.988	0.988
	CIC-ToN-IoT	13	0.993	0.993	0.825	0.873

Quick links

Search