Skip to main content

Table 7 Performance comparisons with relevant systems using gold standard entity annotations on the test dataset of the CDR corpus

From: A document level neural model integrated domain knowledge for chemical-induced disease relations

Methods

System

Methods

Text and concept level

P(%)

R(%)

F(%)

NN with KB

ATT_KB_sum

LSTM+CNN + CTD

Doc_E

60.7

78.7

68.5

LSTM+CNN + CTD + pp

 

63.6

76.8

69.6

Li et al. [17]

CNN

Doc_M

57.8

54.2

55.9

CNN + CTD

 

60.0

81.5

69.1

Verga et al. [16]

Transformer

Doc_E

55.6

70.8

62.1

Transformer+ Extra data

 

64.0

69.2

66.2

Tradional ML with KB

Alam et al. [11]

SVM + CTD + pp

Doc_E + Sen_M

43.7

80.4

56.6

Xu et al. [9]

SVM + CTD + SIDER+MEDI

Doc_E + Sen_M

65.8

68.6

67.2

Pons et al. [8]

SVM + Graph DB

Doc_E

73.1

67.6

70.2

Peng et al. [7]

SVM + CTD + Rules

Doc_E

68.2

66.0

67.1

SVM + CTD + Rules +Extra data

 

71.1

72.6

71.8

Lowe et al. [10]

rules+Ontology+WIKI+PP

Sen_M

59.3

62.3

60.8

NN without KB

Gu et al. [13]

CNN + ME+pp

Doc_M + Sen_M

55.7

68.1

61.3

Zhou et al. [12]

LSTM+SVM + pp

Sen_M

55.6

68.4

61.3

Gu et al. [18]

ME

Doc_M + Sen_M

62.0

55.1

58.3

  1. The 4-th column denotes the text level and the concept level when candidate instances are constructed. “Doc” denotes the document level, “Sen” denotes the sentence level, “_E” denotes entity-based candidate pairs and “_M” denotes mention-based candidate pairs. In addition, all results listed in this table come from the corresponding improved systems after the CDR challenge. The highest F-scores  in each group of methods are highlighted in bold