Table 4. Performance comparison of different methods on the SGCC dataset ( FPR = 0.5%).

Dataset	Model	Precision	Recall	F1-score	Accuracy
SGCC	SVM (Depuru, Wang & Devabhaktuni, 2011)	0.907	0.613	0.731	0.800
	Wide & Deep CNN (Zheng et al., 2017)	0.929	0.812	0.867	0.889
	LSTM (Munawar et al., 2021)	0.932	0.863	0.896	0.911
	CNN-LSTM (Almazroi & Ayub, 2021)	0.934	0.887	0.910	0.922
	Autoencoder (Takiddin et al., 2022)	0.937	0.925	0.931	0.939
	Transformer (Ours)	0.939	0.963	0.951	0.956