Table 9.

The performance of ensemble model classification for the features with kappa >  = 0.2 and all the features (ACC, Prec, Recall, F1, kappa).

Ensemble model	Features with kappa >  = 0.2						All features
	ACC	Prec	Recall	F1	kappa	AUC	ACC	Prec	Recall	F1	kappa	AUC
Scratch CNN	0.77	0.80	0.71	0.75	0.53	0.77	0.74	0.73	0.76	0.75	0.48	0.74
Tunned Vgg16 with data augmentation	0.76	0.82	0.66	0.73	0.52	0.76	0.71	0.70	0.76	0.73	0.43	0.71
Original Vgg16 with data Augmentation	0.63	0.66	0.53	0.59	0.26	0.63	0.62	0.66	0.49	0.57	0.24	0.62
All models ensemble	0.73	0.78	0.66	0.71	0.47	0.73	0.71	0.72	0.68	0.71	0.43	0.71

Bold values indicate best performance of the classifiers.