Skip to main content

Table 1 Extracted information from text content of the test set

From: ChemEx: information extraction system for chemical data curation

 

Exact Matches

Partial Matches

False Positive

False Negative

Precision

Recall

Compounds

203

15

41

105

83.20%

62.85%

Organisms

91

21

3

5

96.81%

77.78%

Assays

80

0

0

15

100.00%

84.21%

  1. The test set consisted of 89 publications with terms "fungus Thailand" from ACS Publications. Only 74 publications reported compounds with 2D chemical structures. Compounds, organisms, and assays were extracted from text content and compared with manually listed entities.