Table 5 The detailed results of the baselines for each type of entity.

From: AgCNER, the First Large-Scale Chinese Named Entity Recognition Dataset for Agricultural Diseases and Pests

Models

Metrics

BIS

OTH

COM

CRO

DIS

DRUG

FER

ORG

PART

PER

PET

PAOG

CUL

BiLSTM-CRF

Precision

91.38

64.51

71.42

95.28

97.86

90.41

85.57

70.99

88.96

89.79

96.65

91.57

88.45

Recall

93.92

52.83

81.89

96.05

97.81

90.74

86.86

73.88

88.83

90.68

96.99

93.21

90.04

F1

92.62

57.69

75.93

95.66

97.83

90.57

86.16

72.37

88.89

90.23

96.82

92.35

89.23

BERTfrozen-BiLSTM-CRF

Precision

91.11

62.01

71.49

96.06

97.57

91.66

85.36

71.92

90.00

90.49

96.77

87.69

89.02

Recall

96.69

58.25

83.38

96.28

97.94

92.56

90.41

80.76

90.77

91.31

97.24

91.79

91.30

F1

93.81

60.03

76.72

96.17

97.75

92.11

87.76

76.07

90.38

90.90

97.00

89.65

90.14