Table 4 Performance comparison of different classifiers
From: Deep-ProBind: binding protein prediction with transformer-based deep learning model
Methods | Dataset | ACC (%) | SN (%) | F1 (%) | SP (%) | MCC |
|---|---|---|---|---|---|---|
LR | Training | 88.48 | 89.03 | 88.03 | 87.89 | 0.769 |
NB | 89.26 | 91.04 | 89.04 | 89.41 | 0.782 | |
RF | 89.91 | 90.65 | 88.65 | 88.59 | 0.795 | |
KNN | 90.25 | 91.36 | 90.36 | 89.28 | 0.812 | |
SVM | 91.15 | 92.42 | 91.47 | 90.01 | 0.830 | |
Deep-ProBind | 92.67 | 93.41 | 93.41 | 91.82 | 0.853 | |
LR | Independent | 91.61 | 92.22 | 92.00 | 91.06 | 0.832 |
NB | 92.01 | 92.86 | 92.85 | 91.22 | 0.840 | |
RF | 92.15 | 92.57 | 92.71 | 91.77 | 0.843 | |
KNN | 92.45 | 93.79 | 91.53 | 91.27 | 0.846 | |
SVM | 92.77 | 94.06 | 91.17 | 91.56 | 0.851 | |
Deep-ProBind | 93.62 | 94.36 | 94.90 | 92.82 | 0.872 |