Skip to main content

Table 6 The effect of using the BANNER entity tagger compared to gold-standard entities on the performance (precision, recall and F-score) of AkanePPI trained on BioInfer.

From: A realistic assessment of methods for extracting gene/protein interactions from free text

 

A

(B)

H

I

L

Precision:

     

With gold-standard entities

29.1

56.8

52.0

66.2

76.7

With BANNER

32.3

49.5

35.1

17.8

50.7

Δ precision

-3.2

7.3

16.9

48.4

26.0

Recall:

     

With gold-standard entities

52.9

85.4

55.8

51.3

40.2

With BANNER

38.9

42.2

37.4

30.1

23.2

Δ recall

14.0

43.2

18.4

21.2

17.0

F-score:

     

With gold-standard entities

37.5

68.2

53.8

57.8

52.8

With BANNER

35.3

45.5

36.2

22.4

31.8

Δ F-score

2.2

22.7

17.6

35.4

21.0

  1. Corpus abbreviations are as follows: A = AIMed; B = BioInfer; H = HPRD50; I = IEPA; L = LLL.