Skip to main content

Advertisement

Table 10 Performance comparison with or without including general terms

From: Identifying named entities from PubMed® for enriching semantic categories

Headword Original evaluation General terms excluded
  Pattern 1 Pattern 2 Pattern 3 Pattern 1 Pattern 2 Pattern 3
Gene 91.0% 76.7% 90.0% 90.0% 73.0% 88.0%
Protein 91.0% 93.0% 91.0% 90.0% 89.0% 85.0%
Disease 93.9% 97.9% 99.0% 93.9% 96.4% 99.0%
Cell 95.2% 98.4% 95.0% 95.2% 71.4% 88.8%
Cells 97.4% 96.7% 98.6% 96.4% 90.0% 95.7%
  1. “General” indicates a term is valid, but the meaning is too general and not useful for enriching SemCat. Scores are the precisions averaged from three reviewers.