Skip to main content

Table 3 Performance on one-word false positives

From: BioCreAtIvE Task1A: entity identification with a stochastic tagger

Occurrences

without post-p

with post-p

corrected

% corrected

0

818

717

101

12.3%

1

162

95

67

41.4%

2+

440

65

375

85.2%

Total

1420

877

543

38.2%

  1. 85% of one-word false positives that correspond to a word, which was seen twice or more times, were corrected with post-processing procedures.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us