Table 11 Performances of the task of named entity recognition in terms of F1-score (and its standard deviation in brackets, for the last setting).
From: PGxCorpus, a manually annotated corpus for pharmacogenomics
Entity matching: (exact or partial) | exact | exact | partial | partial |
|---|---|---|---|---|
Considering hierarchy: (yes or no) | no | yes | no | yes |
Chemical | 76.07 | 76.07 | 82.67 | 82.67 (7.24) |
Genomic_factor | 22.86 | 71.41 | 27.68 | 83.19 (5.90) |
↳Gene_or_protein | 85.72 | 85.72 | 90.58 | 90.05 (3.89) |
↳Genomic_variation | 2.67 | 49.13 | 3.83 | 71.18 (9.55) |
↳Limited_variation | 47.08 | 47.02 | 72.71 | 71.57 (9.50) |
↳Haplotype | 66.97 | 66.97 | 72.47 | 72.47 (19.34) |
Phenotype | 31.76 | 50.80 | 48.48 | 69.57 (5.40) |
↳Disease | 66.90 | 66.88 | 75.68 | 72.59 (7.30) |
↳Pharmacokinetic_phenotype | 29.30 | 29.30 | 36.47 | 36.27 (19.40) |
↳Pharmacodynamic_phenotype | 38.54 | 38.50 | 58.84 | 58.18 (10.11) |
Macro average | 49.15 | 59.11 | 59.76 | 71.93 (5.64) |
