Skip to main content

Table 6 Dictionary matching results on the development corpus

From: Ambiguity and variability of database and software names in bioinformatics

Fold

Recall (%)

Precision (%)

F-score (%)

  1

46 (43)

41 (39)

43 (41)

  2

34 (31)

37 (34)

36 (32)

  3

36 (34)

24 (23)

29 (27)

  4

55 (53)

46 (45)

50 (49)

  5

76 (75)

44 (43)

56 (55)

Min

34 (31)

24 (23)

29 (27)

Max

76 (75)

46 (45)

56 (55)

Mean

49 (47)

38 (37)

43 (41)

  1. Note that for Fold 3, a decrease in score (of about 8 % F-score) is observed if the LINNAEUS abbreviation detected is disabled. Strict scores provided in brackets
  2. P Precision, R Recall, F F-score on the development set using dictionary look-up