Table 7.
Machine learning results with post-processing on the development corpus
| Fold | Recall (%) | Precision (%) | F-score (%) |
|---|---|---|---|
| 1 | 51 (44) | 71 (60) | 59 (51) |
| 2 | 44 (35) | 88 (71) | 59 (47) |
| 3 | 51 (44) | 76 (66) | 61 (53) |
| 4 | 65 (60) | 73 (67) | 69 (63) |
| 5 | 80 (76) | 74 (70) | 77 (73) |
| Min | 44 (35) | 71 (60) | 59 (47) |
| Max | 80 (76) | 88 (71) | 77 (73) |
| Mean | 58 (52) | 76 (67) | 65 (58) |
| Micro Avg | 56 (50) | 76 (67) | 65 (57) |
Strict scores provided in brackets
P Precision, R Recall, F F-score on the development set using machine learning with post-processing (5-cross fold)