Table 2:
Number of entities by type and by train/validation/test split in FRD.
| Entity Type | Train | Val. | Test |
|---|---|---|---|
| treatment | 24,504 | 2,746 | 3,204 |
| chronic_disease | 20,999 | 2,371 | 2,614 |
| upper_bound | 11,136 | 1,228 | 1,371 |
| lower_bound | 10,734 | 1,179 | 1,380 |
| clinical_variable | 10,439 | 1,152 | 1,424 |
| cancer | 7,436 | 826 | 933 |
| gender | 2,830 | 323 | 358 |
| pregnancy | 2,174 | 248 | 315 |
| age | 2,095 | 223 | 262 |
| allergy_name | 1,504 | 183 | 194 |
| contraception_consent | 1,301 | 130 | 169 |
| language_fluency | 391 | 44 | 47 |
| bmi | 235 | 25 | 24 |
| technology_access | 108 | 13 | 11 |
| ethnicity | 64 | 7 | 8 |
| Total | 95,950 | 10,698 | 12,314 |