TABLE 3.
The 10 most frequent tokens, excluding stopwords and punctuation marks, within various window sizes around entities incorrectly labeled by human reviewers.
| Window size = 1 | Window size = 3 | Window size = 5 | ||||
|---|---|---|---|---|---|---|
| # | Token | Count | Token | Count | Token | Count |
| 1 | 300 | 4 | Mg | 19 | mg | 23 |
| 2 | oral | 3 | once/day | 7 | daily | 8 |
| 3 | dose | 3 | treatment | 6 | treatment | 8 |
| 4 | intravenous | 2 | 300 | 5 | once/day | 7 |
| 5 | 500 | 2 | treated | 4 | 300 | 7 |
| 6 | intravenously | 1 | Oral | 4 | oral | 6 |
| 7 | include | 1 | Once | 4 | recipients | 5 |
| 8 | Both | 1 | Dose | 4 | treated | 4 |
| 9 | resistance | 1 | Cidofovir | 3 | twice | 4 |
| 10 | treatment | 1 | resistance | 3 | include | 4 |