Table 8 Comparison results with state-of-arts of the CCKS 2017 datasets.

From: Enhancing medical text classification with GAN-based data augmentation and multi-task learning in BERT

Model

Precision

Recall

F1 score

BERT

87.90

89.23

89.93

RoBERTa27

89.40

89.18

87.90

XLNet28

90.24

88.89

87.39

Wind2vec-BERT29

90.67

89.90

90.89

Bert-pin30

92.14

89.23

91.54

Mask-guided BERT11

93.12

91.34

91.23

Proposed

94.52

93.24

91.80

Back to article page