Skip to main content

Table 1 Performance comparison on maximum F1 score with existing models

From: Relation extraction between bacteria and biotopes from biomedical texts with attention mechanisms and domain-specific contextual representations

Methods

Team

F

R

P

SVM-based

TEES [5]

42.27

38.35

61.61

 

VERSE [6]

55.80

61.50

51.00

DL-based

DUTIR [15]

47.80

39.70

60.00

 

TurkuNLP [16]

52.10

44.80

62.30

 

DET-BLSTM [17]

57.14

57.99

56.32

 

BGRU-Attn [18]

57.42

69.82

48.76

 

Our model

60.77

65.28

56.85

  1. The listed results derive from the corresponding papers. F: F1 score; R: recall; P: precision. Our model used all of the proposed contributions (the results from the last row in Table 6). The highest scores are highlighted in bold