Research on Chinese EMR named entity recognition based on RoBERTa-WWM-BiLSTM-CRF
Zhang Fangcong,Qin Qiuli,Jiang Yong,Zhuang Runtao
(School of Economics and Management , Beijing Jiaotong University, Beijing 100044, China)
(National Clinical Medical Research Center for nervous system diseases, Beijing Tiantan Hospital Affiliated to Capital Medical University, Beijing 100050, China)
(Community health service center of Beijing Jiaotong University, Beijing 100044, China)
[Objective] Aiming at the problems of polysemy and incomplete word recognition in Chinese EMR entity recognition.
[Methods] The deep learning model RoBERTa-WWM-BiLSTM-CRF is used to improve the effect of named entity recognition of Chinese electronic medical record. Four groups of experiments are used to compare and analyze the influence of different models on the effect of Chinese electronic medical record entity recognition.
[Results] The results show that the F1 value of the model is 89.08%.
[Limitations] The data set used is small, and the entity recognition effect of some departments is relatively general. For example, the F1 value of respiratory department is only 81.11%. [Conclusions] Experiments show that the RoBERTa-WWM-BiLSTM-CRF model constructed in this paper is more suitable for the task of Chinese electronic medical record named entity recognition, and effectively solves the problems of polysemy and incomplete word recognition in Chinese electronic medical record named entity recognition. The F1 value of the model reaches 89.08%.
张芳丛, 秦秋莉, 姜勇, 庄润涛. 基于RoBERTa-WWM-BiLSTM-CRF的中文电子病历命名实体识别研究
[J]. 数据分析与知识发现, 0, (): 1-.
Zhang Fangcong, Qin Qiuli, Jiang Yong, Zhuang Runtao. Research on Chinese EMR named entity recognition based on RoBERTa-WWM-BiLSTM-CRF
. Data Analysis and Knowledge Discovery, 0, (): 1-.
