Fine-tuning BERT for Joint Entity and Relation Extraction in Chinese Medical Text
This work addresses the problem of structuring medical text for healthcare professionals, but it is incremental as it builds on existing BERT-based methods with a specific attention mechanism.
The paper tackled joint entity and relation extraction in Chinese medical text by integrating BERT with a dynamic range attention mechanism, achieving F1-scores of 96.89% for named entity recognition and 88.51% for relation classification, outperforming state-of-the-art methods by 1.65% and 1.22%, respectively.
Entity and relation extraction is the necessary step in structuring medical text. However, the feature extraction ability of the bidirectional long short term memory network in the existing model does not achieve the best effect. At the same time, the language model has achieved excellent results in more and more natural language processing tasks. In this paper, we present a focused attention model for the joint entity and relation extraction task. Our model integrates well-known BERT language model into joint learning through dynamic range attention mechanism, thus improving the feature representation ability of shared parameter layer. Experimental results on coronary angiography texts collected from Shuguang Hospital show that the F1-score of named entity recognition and relation classification tasks reach 96.89% and 88.51%, which are better than state-of-the-art methods 1.65% and 1.22%, respectively.