Table 16.

Comparison with other works.

Dataset	Reference	Year	Analysis	No. of Classes	Method	Accuracy (%)
CMD	Mahdavifar et al. [17]	2020	Dynamic	5	PLDNN	97.84
	Mohamed et al. [45]	2021	Static	2	NB, SVM, KNN, DT	86
	Ahmed et al. [54]	2022	Hybrid	5	RF, MLP	97.5
	Musikawan et al. [40]	2022	Static	2	DNN	98.18
			Dynamic	2	DNN	93.5
	Ullah et al. [46]	2022	Dynamic	4	NB, SVM, DT, LR, RF	99.11
	Padmavathi et al. [55]	2022	Dynamic	5	K-means, PCA	88
	Jundi et al. [47]	2023	Hybrid	5	XGBoost, GE	98
	Tang et al. [49]	2024	Hybrid	2	HBI, DNN-AM	98.67
	Proposed	2024	Dynamic	5	IZOA-LightGBM	99.75
CCA	Rahali et al. [37]	2020	Dynamic	12	Semi-Supervised Deep Learning	93.36
	Batouche et al. [56]	2021	Static	14	RF	89
	Musikawan et al. [40]	2022	Static	2	DNN	97.72
			Dynamic	14	DNN	78.82
	Al-Andoli et al. [50]	2022	Static	12	PDL-FEMC	97.6
	Xie et al. [48]	2023	Static	15	MLD-Model	83.17
	Islam et al. [57]	2023	Dynamic	12	Ensemble ML	95
	Li et al. [51]	2024	Static	12	SynDroid-RF	94.31
	Huang et al. [39]	2024	Hybrid	15	RF	88.2
	Proposed	2024	Dynamic	12	IZOA-LightGBM	98.86
AAGM	Bovenzi et al. [58]	2022	Dynamic	3	RF	97
	Alani et al. [52]	2022	Dynamic	2	AdStop	97.08
	Ullah et al. [53]	2024	Dynamic	3	FL	93.85
	Proposed	2024	Dynamic	3	IZOA-LightGBM	97.79

Bold indicates the highest accuracy.