Thyroid Discharge Summary NER Based on BERT-BiLSTM-CRF Model
Yue Li, Xian Jin Xie, Dehua Chen, Anxing Wen, Ang Li, Zunli Xie · 2022 3rd International Conference on Electronic Communication and Artificial Intelligence (IWECAI) · 2022
The text study of Chinese thyroid discharge summary plays an important role in predicting thyroid disease. The results of named entity recognition are crucial for later text structuring. In recent common sequence annotation tasks, the contextual semantic issues of words in the text cannot be captured because the encoding of words is static, which can lead to a poor entity recognition rate. In this paper, a method combining Bert model with BiLstm-CRF is proposed to achieve a relatively higher recognition rate.