Table 4 Comparison of the STSS algorithm with traditional self-training on the edge detection dataset.
From: Self-training in significance space of support vectors for imbalanced biomedical event data
Class label | Precision (%) | Precision (%) | F-measure (%) | |||
|---|---|---|---|---|---|---|
| Â | Self-training | STSS | Self-training | STSS | Self-training | STSS |
AtLoc | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
Cause | 47.36 | 49.21 | 43.54 | 43.61 | 45.36 | 46.24 |
Cause-Theme | 76.2 | 67.85 | 53.72 | 80.6 | 63.01 | 73.67 |
Site | 29.82 | 29.25 | 33.2 | 33.0 | 31.4 | 31.01 |
Theme | 57.11 | 57.72 | 90.33 | 90.92 | 69.97 | 70.61 |
ToLoc | 48.42 | 48.94 | 96.42 | 96.68 | 64.46 | 64.98 |
Negative | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |