Table 2 Training cohort and test cohort distributions
From: Using natural language processing and machine learning to identify breast cancer local recurrence
| Â | Total | Local Recurrence | Percentage (%) | Overall percentage (%) |
|---|---|---|---|---|
Double annotation set | 701 | 193 | 27.53% | 8.13% |
 Cross-validation set | 490 | 143 | 29.18% | |
 Held-out test set | 211 | 50 | 23.70% | |
Single annotation set | 6198 | 368 | 5.94% |