Table 6 Comparison of Post-Hoc detection with our method.

method	Precision	Accuracy	F1_score	AUC	Recall
StackedCNN-attentionalBiGRUmodel³⁶	0.98	0.92	0.96	0.95	0.95
FCVM³⁷	0.91	0.90	0.85	0.86	0.85
LSTM³⁸	0.90	0.86	0.83	0.85	0.86
ML³⁹	0.88	0.85	0.91	0.72	0.87
DL-BERT⁴⁰	0.82	0.75	0.84	0.90	0.90
CBSigIDS⁴¹	0.85	0.84	0.83	0.81	0.82
TLDANN⁴²	0.87	0.88	0.88	0.85	0.87
LSTMandstackedGRU-basedattentionmodel⁴³	0.89	0.91	0.93	0.92	0.91
LSTM⁴⁴	0.88	0.87	0.91	0.79	0.88
Cyber-PersonaIdentification⁴⁵	0.81	0.85	0.89	0.88	0.79
Fine-grainedapproach⁴⁶	0.84	0.88	0.90	0.78	0.88
OriginalTransformer⁴⁷	0.98	0.93	0.89	0.97	0.81
DistilledTrans⁴⁷	0.97	0.94	0.91	0.97	0.84
BERT+FL⁴⁷	0.93	0.96	0.96	0.96	0.95
Roberta+FL⁴⁷	0.95	0.97	0.94	0.96	0.94
Hybrid Learning Approach⁴⁷	0.85	0.88	0.85	0.84	0.84
Transformer+TCN	0.49	0.97	0.65	0.984	0.95

Quick links

Search