Skip to main content

Table 2 Composition of species tags in the manually annotated corpus and false negative predictions by LINNAEUS relative to the manually annotated corpus on the same document set

From: LINNAEUS: A species name identification system for biomedical literature

Category

Number of tags in corpus

Number of false negatives

Misspelled

46

11

Incorrect case

130

128

OCR/technical errors

18

16

Enumeration

2

1

Incorrectly used name

79

66

Modifier

1,217

125

Normal mention

2,788

12

  1. A detailed description of the different tag categories is provided in the Methods.