Table 6 Identification results of MCSR-NER-model.

From: Automatic construction of risk transmission network about subway construction based on deep learning models

Testing set

Processed batches: 14,084, pre-labeled phrases: 656 ; identify: 669; Correct: 520

Entities types

Precision

Recall

F1 value

Number of identified entities

Total

77.73%

79.27%

78.49%

669

EF

53.85%

61.76%

57.53%

39

HF

77.78%

83.33%

80.46%

45

RE

84.58%

85.94%

85.26%

441

MD

68.18%

65.22%

66.67%

22

TM

62.30%

61.79%

62.04%

122

Back to article page