计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (9): 287-294.doi: 10.11896/jsjkx.220900226

• 人工智能 • 上一篇    下一篇

融合机器阅读理解的中文医学命名实体识别方法

罗媛媛1, 杨春明1,3, 李波1, 张晖2, 赵旭剑1,3   

  1. 1 西南科技大学计算机科学与技术学院 四川 绵阳 621000
    2 西南科技大学数理学院 四川 绵阳 621000
    3 四川省大数据与智能系统工程技术研究中心 四川 绵阳 621010
  • 收稿日期:2022-09-23 修回日期:2022-12-02 出版日期:2023-09-15 发布日期:2023-09-01
  • 通讯作者: 杨春明(yangchunming@swust.edu.cn)
  • 作者简介:(2306543568@qq.com)
  • 基金资助:
    四川省科技厅重点研发项目(2021YFG0031);四川省省级科研院所科技成果转化项目(22YSZH0021)

Chinese Medical Named Entity Recognition Method Incorporating Machine ReadingComprehension

LUO Yuanyuan1, YANG Chunming1,3, LI Bo1, ZHANG Hui2, ZHAO Xujian1,3   

  1. 1 School of Computer Science and Technology,Southwest University of Science and Technology,Mianyang,Sichuan 621000,China
    2 School of Mathematics and Physics,Southwest University of Science and Technology,Mianyang,Sichuan 621000,China
    3 Sichuan Big Data and Intelligent System Engineering Technology Research Center,Mianyang,Sichuan 621010,China
  • Received:2022-09-23 Revised:2022-12-02 Online:2023-09-15 Published:2023-09-01
  • About author:LUO Yuanyuan,born in 1998,postgra-duate,is a member of China Computer Federation.Her main research interests include knowledge graphs and natural language processing.
    YANG Chunming,born in 1980,asso-ciate professor,is a member of China Computer Federation.His main research interests include nature language processing and machine learning.
  • Supported by:
    Key R & D Project of Science & Technology Department of Sichuan Province(2021YFG0031) and Scientific and Technological Achievements Transformation Project of Sichuan Provincial Scientific Research Institute(22YSZH0021).

PDF (PC)

2056

摘要： 医学命名实体识别是自动构建大规模医学知识库的关键,但医学文本中存在实体嵌套现象,采用序列标注的方法不能识别出嵌套中的实体。文中提出了基于阅读理解框架的中文医学命名实体识别方法,该方法将嵌套命名实体识别问题建模为机器阅读理解问题,使用BERT建立阅读理解问题和医学文本之间的联系,并引入多头注意力机制强化问题和嵌套实体之间的语义联系,最后用两个分类器对实体开头和结尾位置进行预测。与目前5种主流方法相比,该方法取得了最优结果,综合F1值达到了67.65%;与经典的实体识别模型BiLSTM-CRF相比,F1值提升了7.17%,其中嵌套较多的临床表现实体提升16.81%。

关键词: 命名实体识别, 中文医学, 嵌套实体, 机器阅读理解, 多头注意力机制

Abstract: Medical named entity recognition is the key to automatically build a large-scale medical knowledge base.However,medical entities are often nested,and it can not be recognized by the sequence labeling method.This paper proposes a Chinese medical named entity recognition method based on reading comprehension framework.It models the nested named entity recognition problem as a machine reading problem,uses BERT to establish the connection between the reading comprehension problem and medical text,and introduces a multi-head attention mechanism to strengthen the semantic connection between the problem and nested named entity,and finally uses two classifiers to predict the beginning and end positions of entities.This method achieves the best results with an F1-score of 67.65% when compared with the current five mainstream methods.Compared with the most classical BiLSTM-CRF,the F1-score improves by 7.17%,and the nested “symptom” entities increase by 16.81%.

Key words: Named entity recognition, Chinese medical, Nested entities, Machine reading comprehension, Multi-head attention mechanism

中图分类号: 

  • TP391.1
[1]CUI Y,CHE W,LIU T,et al.Revisiting Pre-Trained Models for Chinese Natural Language Processing[C]//Empirical Methods in Natural Language Processing.2020.657-668.
[2]VASWANI A,SHAZEER N,PARMAR N,et al.Attention is allyou need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems.2017:6000-6010.
[3]MORWAL S,JAHAN N,CHOPRA D.Named entity recogni-tion using hidden Markov model [J].International Journal on Natural Language Computing,2012,4(1):15-23.
[4]JU Z,WANG J,ZHU F.Named entity recognition from biome-dical text using SVM[C]//2011 5th International Conference on Bioinformatics and Biomedical Engineering.IEEE,2011:1-4.
[5]SONG S,ZHANG N,HUANG H.Named entity recognitionbased on conditional random fields[J].Cluster Computing,2019,22(3):5195-5206.
[6]GUI T,MA R,ZHANG Q,et al.CNN-Based Chinese NER with Lexicon Rethinking[C]//International Joint Conference on Artificial Intelligence.Macao,China,2019:4982-4988.
[7]CHOWDHURY S,DONG X,QIAN L,et al.A multitask bi-directional RNN model for named entity recognition on Chinese electronic medical records[J].BMC Bioinformatics,2018,19(17):75-84.
[8]OUYANG E,LI Y,JIN L,et al.Exploring n-gram characterpresentation in bidirectional RNN-CRF for Chinese clinical named entity recognition[C]//CEUR Workshop Proceedings.2017:37-42.
[9]XU K,ZHOU Z,HAO T,et al.A bidirectional LSTM and conditional random fields approach to medical named entity recognition[C]//International Conference on Advanced Intelligent Systems and Informatics.Cham:Springer,2017:355-365.
[10]HUANG Z,XU W,YU K.Bidirectional LSTM-CRF models for sequence tagging[J].arXiv:1508.01991,2015.
[11]TANG B,WANG X,YAN J,et al.Entity recognition in Chinese clinical text using attention-based CNN-LSTM-CRF[J].BMC Medical Informatics and Decision Making,2019,19(3):89-97.
[12]MIKOLOV T,CHEN K,CORRADO G,et al.Efficient estimation of word representations in vector space[J].arXiv:1301.3781,2013.
[13]DEVLIN J,CHANG M W,LEE K,et al.Bert:Pre-training of deep bidirectional transformers for language understanding[J].arXiv:1810.04805,2018.
[14]DAI Z,WANG X,NI P,et al.Named entity recognition usingBERT BiLSTM CRF for Chinese electronic health records[C]//2019 12th International Congresson Image and Signal Proces-sing,Biomedical Engineering and Informatics(CISP-BMEI).2019:1-5.
[15]LI X,ZHANG H,ZHOU X H.Chinese clinical named entityrecognition with variant neural structures based on BERT methods[J].Journal of Biomedical Informatics,2020,107:103422.
[16]JU M,MIWA M,ANANIADOU S.A neural layered model for nested named entity recognition[C]//Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies.New Orleans,Louisiana,2018:1446-1459.
[17]XU H,LIU H,JIA Q,et al.A nested named entity recognitionmethod for traditional Chinese medicine records[C]//International Conference on Artificial Intelligence and Security.Cham:Springer,2021:488-497.
[18]ZHENG C,CAI Y,XU J,et al.A Boundary-aware Neural Model for Nested Named Entity Recognition[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing(EMNLP-IJCNLP).2019:357-366.
[19]SUN L,SUN Y,JI F,et al.Joint Learning of Token Context and Span Feature for Span-Based Nested NER[J].IEEE/ACM Transactions on Audio,Speech,and Language Processing,2020,28:2720-2730.
[20]MARINHO Z,MENDES A,MIRANDA S,et al.Hierarchicalnested named entity recognition[C]//Proceedings of the 2nd Clinical Natural Language Processing Workshop.2019:28-34.
[21]WANG B,LU W,WANG Y,et al.A Neural Transition-basedModel for Nested Mention Recognition[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing.2018:1011-1017.
[22]LU W,ROTH D.Joint mention extraction and classificationwith mention hypergraphs[C]//Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing.Lisbon,Portugal,2015:857-867.
[23]KATIYAR A,CARDIE C.Nested Named Entity RecognitionRevisited[C]//Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies,Volume 1(Long Papers).2018:861-871.
[24]LEVY O,SEO M,CHOI E,et al.Zero-shot relation extraction via reading comprehension[J].arXiv:1706.04115,2017.
[25]LIU J,CHEN Y,LIU K,et al.Event extraction as machinereading comprehension[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing.2020:1641-1651.
[26]LIU S,ZHANG X,ZHANG S,et al.Neural machine reading comprehension:Methods and trends[J].Applied Sciences,2019,9(18):3698.
[27]CAO J,ZHOU X,XIONG W,et al.Electronic Medical Record Entity Recognition via Machine Reading Comprehension and Biaffine[J].Discrete Dynamics in Nature and Society,2021,2021(9):16408371-1-16408371-8.
[28]CHIANG Y L,LIN C H,SUNG C L,et al.Nested Named Entity Recognition for Chinese Electronic Health Records with QA-based Sequence Labeling[C]//Proceedings of the 33rd Confe-rence on Computational Linguistics and Speech Processing.2021:18-25.
[29]LI X,FENG J,MENG Y,et al.A unified MRC framework fornamed entity recognition[J].arXiv:1910.11476,2019.
[30]YANG P,CONG X,SUN Z,et al.Enhanced Language Representation with Label Knowledge for Span Extraction[J].arXiv:2111.00884,2021.
[31]MNIH V,HEESS N,GRAVES A.Recurrent models of visual attention[J].Advances in Neural Information Processing Systems,2014,27:2204-2212.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发