Network Intrusion Detection Method Based on FCWGAN and BiLSTM

Table 2

Model parameter settings.

Parameter	Setting
XGBoost maximum depth	12
XGBoost gamma value	0
CWGAN learning rate	0.0001
CWGAN training iterations	200
Noise dimension	32
Batch size setting	1024
Loss function	Categorical cross-entropy
Optimizer	Nadam
Optimizer learning rate	0.001
BiLSTM cell count	64/128
Dropout rate	0.5