Skip to main content

Table 7 Performance comparison with different LMs as encoder in our method

From: BioEGRE: a linguistic topology enhanced method for biomedical relation extraction based on BioELECTRA and graph pointer neural network

Language model

Precision (%)

Recall (%)

F1-score (%)

BioBERT + GPNN

76.47

78.03

77.24

SciBERT + GPNN

71.86

77.78

74.70

PubMedBERT + GPNN

70.6

74.5

72.5

ClinicalBERT + GPNN

75.90

77.02

76.46

Proposed (BioELECTRA + GPNN)

77.97

82.07

79.97

  1. The bold indicates list the results of the models represents the best among the results
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us