Table 1 Difference in the number of PubTator predictions in a general set of PubMed articles, selected to be moderately enriched in chemical mentions, and the articles selected for annotation in the NLM-Chem corpus.
From: NLM-Chem, a new resource for chemical entity recognition in PubMed full text literature
|  | Annotations per abstract in 20 K PubMed articles | Annotations per abstract in NLM-Chem corpus |
|---|---|---|
Chemicals | 3.98 | 14.58 |
Species | 6.23 | 3.45 |
Gene | 5.91 | 6.29 |
Disease | 5.07 | 5.69 |
Mutation | 0.06 | 0.37 |
