Skip to main content

Table 6 Performance (in %) of individual systems in combination with NP chunking, before and after stop-word removal.

From: Recognition of chemical entities: combining dictionary-based and grammar-based approaches

  Baseline + NP chunking Baseline + NP chunking + stop-words
  Precision Recall F-score Precision Recall F-score
ChEBI 56.3 39.4 46.4 77.5 38.5 51.5
ChEMBL 87.8 18.2 30.1 88.6 18.2 30.1
ChemSpider 70.1 37.9 49.2 81.5 37.3 51.2
DrugBank 62.9 16.8 26.5 76.6 16.7 27.5
HMDB 73.5 33.7 46.2 82.0 33.1 47.2
NPC 46.8 26.0 33.5 59.1 25.7 35.9
TTD 66.6 14.4 23.6 83.0 14.0 24.0
PubChem 32.7 57.0 41.6 61.5 55.9 58.6
Jochem 64.3 50.6 56.7 67.4 50.6 57.8
UMLS 36.6 49.2 42.0 44.3 48.9 46.5
ChEBI Family 18.4 15.9 17.1 28.8 15.6 20.3