Skip to main content

Table 5 Comparison of BioNER for genes and proteins

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

Method\dataset

BC2GM

JNLPBA

p

r

f1

p

r

f1

BiLSTM-CRF [28]

81.57

79.48

80.51

71.35

75.74

73.48

CollaboNet [29]

80.49

78.99

79.73

74.43

83.22

78.58

MTM-CW [31]

82.10

79.42

80.74

70.91

76.34

73.52

BioKMNER [32]

-

-

84.92

-

-

77.72

BioBERT-MRC [33]

87.04

83.98

85.48

75.96

82.13

78.93

MTL-LS [34]

–

–

82.92

–

–

–

BioELECTRA[35]

–

–

84.69

–

–

80.07

BioBERT [11]

84.32

85.12

84.72

72.24

83.56

77.49

Proposed

84.97

85.32

85.15

72.69

84.54

78.16

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