Table 1 GM results: performances of systems and individual components on the test and training data.
From: Concept recognition for extracting protein interaction relations from biomedical text
| Â | Test Data | Training Data | ||||
|---|---|---|---|---|---|---|
Tagger | Precision | Recall | F-measure | Precision | Recall | F-measure |
CCP | 77.30 | 77.74 | 77.52 | 83.68 | 83.48 | 83.58 |
ABNER | 80.38 | 73.26 | 76.65 | 83.85 | 80.86 | 82.33 |
LingPipe | 72.53 | 80.00 | 76.09 | 88.47 | 92.77 | 90.57 |
2/3 Majority | 85.54 (2) | 76.83 (3) | 80.95 (3) | 91.15 | 86.33 | 88.68 |
Unanimous | 92.78 (1) | 49.12 (4) | 64.24 (4) | 94.56 | 61.41 | 74.46 |
Overlap | 66.22 (4) | 83.72 (2) | 73.94 (4) | 79.27 | 91.17 | 84.80 |
Median | 85.08 | 79.05 | 81.32 | Â | Â | Â |