Table 6.
Aggregate statistics for token-based evaluation of all submissions – all PHI categories
| Minimum | Mean | Median | Maximum | Std. deviation | |
|---|---|---|---|---|---|
| Micro Precision | 0.716 | 0.927 | 0.953 | 0.982 | 0.073 |
| Micro Recall | 0.211 | 0.777 | 0.863 | 0.941 | 0.203 |
| Micro F1 | 0.344 | 0.832 | 0.907 | 0.961 | 0.164 |
| Macro Precision | 0.731 | 0.924 | 0.951 | 0.981 | 0.069 |
| Macro Recall | 0.244 | 0.772 | 0.856 | 0.940 | 0.203 |
| Macro F1 | 0.385 | 0.828 | 0.902 | 0.960 | 0.161 |