Skip to main content

Table 4 Comparison of BioNER for diseases

From: BioByGANS: biomedical named entity recognition by fusing contextual and syntactic features through graph attention network in node classification framework

Method\Dataset

NCBI-disease

BC5CDR-disease

p

r

f1

p

r

f1

TaggerOne [27]

85.10

80.80

82.90

85.20

80.20

82.60

BiLSTM-CRF [28]

86.11

85.49

85.80

87.60

86.25

86.92

CollaboNet [29]

85.61

82.61

84.08

85.61

82.61

84.08

MTM-CW [31]

85.86

86.42

86.14

89.10

88.47

88.78

DABLC [30]

88.30

89.01

88.60

89.10

87.50

88.30

BioKMNER [32]

–

–

90.08

–

–

–

BioBERT-MRC [33]

89.67

90.42

90.04

88.61

87.07

87.83

MTL-LS [34]

–

–

89.25

–

–

87.28

BioELECTRA[35]

–

–

89.38

–

–

85.84

BioBERT [11]

88.22

91.25

89.71

86.47

87.84

87.15

Proposed

89.99

93.20

91.57

86.69

88.82

87.74

  1. Bold indicates the best performances of models in each subtask