Table 6

Results from Steps 1-5a and 5b, and the Evaluation Steps 1 and 2.

Regular expression method

Term voting method


TOTAL terms

98435

100%

98435

100%

Selected set

13755

14%

13755

14%

Excluded set

84680

86%

84680

86%

Sample of excluded

3140

100%

Wrong (false negative)

49

1.6%

Correct (true negative)

3091

98.4%

Proportionate number of bona fide terms in excluded set

1321

Sample of included

2070

100%

2287

100%

Wrong (false positive)

1538

74.3%

1974

86.3%

Correct (true positive)

532

25.7%

313

13.7%

Probable number of bona fide terms in selected set

3535

1883

Recall

0.728

Precision

0.257

0.137


Brewster et al. BMC Bioinformatics 2009 10(Suppl 5):S1   doi:10.1186/1471-2105-10-S5-S1

Open Data