Table 6. Summary table with performance metrics reported as median (95% CI) and F 1 interquartile variance (IQV) after 200 bootstrap iterations.
The performance metrics are compared across pipelines using BoW together with distributed representations.
| Pipeline | Precision (%) | Recall (%) | F 1 (%) | F 1 IQV |
|---|---|---|---|---|
| Unigrams | 80.1 (73.9,86.0) | 82.3 (74.1,88.6) | 80.6 (75.8,85.2) | 9.4 |
| Unigrams + BioBERT mean pooling | 83.7 (76.7,89.1) | 80.4 (74.1,87.3) | 81.7 (77.8,86.0) | 8.2 |
| Unigrams + BioBERT mean + min&max pooling | 83.8 (75.6,88.8) | 79.1 (73.4,85.4) | 81.0 (77.2,85.4) | 8.2 |