Skip to main content

Table 3 The 20 n-grams that are the only common candidate terms among the 880 term pairs from both VetCN and PMSB datasets

From: Exploring semantic deep learning for building reliable and reusable one health knowledge from PubMed systematic reviews and veterinary clinical notes

  1. The character ‘|’ that appears in the first column separates the different neural language models. The grey background indicates that the target term and the candidate term has the same focus concept, i.e. same UMLS CUI