Table 2.

Classification accuracy of $\text{NeurDNet}$ in the two cases of employing binary and probabilistic features.

Classifier	Binary features						Probabilistic features
	25%	35%	45%	55%	65%	75%	25%	35%	45%	55%	65%	75%
RF (entropy)	85.69	84.24	82.91	81.94	82.43	78.68	86.18	85.43	83.79	82.66	82.20	78.21
RF (gini)	85.43	84.59	83.43	82.35	81.97	78.28	86.49	84.81	84.27	82.63	82.57	78.29
SVM (rbf)	85.68	84.65	84.24	82.19	83.10	79.46	86.33	85.83	85.38	82.09	82.68	79.01
SVM (linear)	84.26	82.69	82.08	81.34	80.78	78.02	85.83	84.77	83.60	82.36	82.02	78.57
NB	83.70	83.55	80.23	81.44	81.67	77.31	85.98	86.42	84.94	83.94	84.15	81.48
LR	85.76	84.41	84.09	83.10	82.83	79.49	87.29	86.10	85.28	83.65	83.38	79.74
AdaBoost	83.97	81.61	80.99	79.95	79.30	75.80	85.03	82.97	81.53	80.01	78.12	73.32
LDA (svd)	79.54	76.25	75.83	73.79	66.21	67.44	77.81	76.41	76.56	72.31	65.12	63.62
LDA (lsqr)	79.54	76.25	75.80	73.77	63.40	49.57	77.81	76.41	76.56	72.31	65.12	49.50
QDA	81.85	83.18	78.69	72.08	63.26	58.62	95.55	93.89	81.73	73.48	56.29	53.13
DT (entropy)	81.21	78.45	77.66	77.63	76.02	74.75	80.40	79.01	77.11	77.57	75.06	71.73
DT (gini)	80.45	80.16	78.51	77.25	77.32	75.25	77.99	78.29	76.89	76.35	74.29	71.84
MLP (10)	85.01	82.40	82.05	81.25	79.79	77.53	84.33	83.03	81.64	80.25	80.04	77.04
MLP (30)	84.64	82.84	82.02	80.85	79.63	77.49	84.53	82.80	81.79	80.50	80.33	77.45

The classification accuracy is measured across different choices of the second-stage classifier, including random forests (RF), support vector machines (SVM), Naive Bayes Classifier (NB), logistic regression (LR), AdaBoost classifier (AB), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), decision trees (DT), and multi layer perceptron (MLP).