Predicting Protein Interactions Using a Deep Learning Method-Stacked Sparse Autoencoder Combined with a Probabilistic Classification Vector Machine

Table 5

Performance comparison of different methods on the human dataset.

Model	Accu (%)	Sens (%)	Prec (%)	MCC (%)
LDA + RF [70]	96.40	94.20	N/A	92.80
LDA + RoF [70]	95.70	97.60	N/A	91.80
LDA + SVM [70]	90.70	89.70	N/A	81.30
AC + RF [70]	95.50	94.00	N/A	91.40
AC + RoF [70]	95.10	93.30	N/A	91.10
AC + SVM [70]	89.30	94.00	N/A	79.20
Proposed method	97.71	91.87	98.10	93.59