Table A.9.
Precision [P.], recall [R.], F1 score [F1] using ensemble classification through a weighted vote of classifiers for dimensionality reduction techniques under optimal settings.
| Dimensionality reduction technique | Total # of features | P. | R. | F1 |
|---|---|---|---|---|
|
| ||||
| Full feature space | 45,695 | 0.738 | 0.818 | 0.776 |
| PCA | 333 | 0.724 | 0.814 | 0.766 |
| Mutual Information | 2,284 | 0.741 | 0.798 | 0.769 |
| Ontology | 32,180 | 0.774 | 0.81 | 0.792 |
| Ontology + PCA | 289 | 0.723 | 0.798 | 0.759 |
| Ontology + MI | 1,605 | 0.749 | 0.798 | 0.773 |
Note: Performance of shaded region is not significantly different from the top overall model (i.e., F1 of 77.4%).