Table 13. The best FC learning optimizer for each combination method and its average classification performance (using Recall metric).

	PCA-FC	FA-FC	FastICA-FC	NCA-FC	MRMR-FC	CHI2-FC	FL-FC
AlexNet	0.85	0.83	0.87	0.89	0.97	0.90	0.96
	RMSP	SGDM	ADAM	ADAM	ADAM	ADAM	ADAM
DenseNet201	0.89	0.88	0.74	0.86	0.84	0.80	0.93
	RMSP	RMSP	ADAM	RMSP	SGDM	RMSP	SGDM
Inception-ResNetv2	0.73	0.82	0.75	0.76	0.76	0.72	0.89
	ADAM	SGDM	ADAM	RMSP	RMSP	ADAM	RMSP
Inceptionv3	0.84	0.75	0.64	0.85	0.84	0.84	0.89
	RMSP	SGDM	ADAM	RMSP	RMSP	ADAM	ADAM
MobileNetv2	0.88	0.81	0.79	0.87	0.87	0.90	0.95
	RMSP	SGDM	ADAM	RMSP	RMSP	RMSP	RMSP
ResNet18	0.77	0.72	0.62	0.73	0.76	0.73	0.76
	SGDM	ADAM	ADAM	ADAM	RMSP	RMSP	SGDM
ResNet50	0.87	0.83	0.71	0.89	0.82	0.82	0.97
	ADAM	RMSP	ADAM	RMSP	RMSP	RMSP	ADAM
ResNet101	0.83	0.92	0.68	0.88	0.79	0.85	0.94
	ADAM	SGDM	ADAM	ADAM	ADAM	RMSP	ADAM
VGG16	0.83	0.87	0.82	0.96	0.96	0.93	0.92
	ADAM	RMSP	ADAM	ADAM	ADAM	ADAM	ADAM
VGG19	0.86	0.89	0.78	0.97	0.96	0.90	0.96
	RMSP	RMSP	ADAM	ADAM	ADAM	ADAM	ADAM
Xception	0.85	0.89	0.74	0.76	0.84	0.85	0.86
	ADAM	ADAM	ADAM	ADAM	RMSP	RMSP	ADAM