Research Article
Conformational B-Cell Epitopes Prediction from Sequences Using Cost-Sensitive Ensemble Classifiers and Spatial Clustering
Table 7
Performance of different ensemble methods on bound and unbound datasets.
| Method | Bound dataset | Unbound dataset | | ACC | SN | SP | | ACC | SN | SP |
| Direct Combination | 0.016 | 0.939 | 0.008 | 0.999 | 0.021 | 0.907 | 0.011 | 0.999 | EasyEnsemble | 0.168 | 0.883 | 0.196 | 0.927 | 0.208 | 0.818 | 0.256 | 0.876 | BalanceCascade | 0.224 | 0.870 | 0.314 | 0.904 | 0.245 | 0.808 | 0.335 | 0.856 | SMOTEBoost | 0.252 | 0.852 | 0.415 | 0.879 | 0.239 | 0.750 | 0.423 | 0.784 | Cost-Sensitive | 0.330 | 0.750 | 0.648 | 0.760 | 0.312 | 0.689 | 0.704 | 0.687 |
|
|