Skip to main content

Table 10 Lemmatization performance of the BioLemmatizer resources on OED set

From: BioLemmatizer: a lemmatization tool for morphological processing of biomedical text

 

Recall

Precision

F-score

Base (MorphAdorner lexicon)

53.34% (431/808)

53.34% (431/808)

53.34%

Base + GENIA

52.97% (428/808)

52.97% (428/808)

52.97%

Base + BioLexicon

54.08% (437/808)

54.08% (437/808)

54.08%

Entire Lexicon

54.21% (438/808)

54.21% (438/808)

54.21%

Rule Only

66.96% (541/808)

66.96% (541/808)

66.96%

Rule + Lexicon Validation

71.29% (576/808)

71.29% (576/808)

71.29%