基于多重注意力机制的中文医疗实体识别

doi:10.19678/j.issn.1000-3428.0064583

摘要/Abstract

摘要： 医疗实体识别是从医疗文本中识别疾病、症状、药物等多种类型的医疗实体，能够为知识图谱、智慧医疗等下游任务的发展提供支持。针对现有命名实体识别模型提取语义特征较单一、对医疗文本语义理解能力不足的问题，提出一种基于多重注意力机制的神经网络模型MANM。为捕获文本中更丰富的语义特征，在模型输入中引入医疗词汇先验知识，通过自注意力机制获取医疗文本的全局语义特征，并利用双线性注意力机制获取词汇和字符层面的潜在语义特征，得到包含字词间依赖关系的特征向量。为提高模型的上下文信息捕捉能力，采用改进的长短时记忆网络提取文本时序特征，同时设计多头自注意力机制获取词语间隐含的关联语义特征。最后融合上述多层次语义特征，利用条件随机场进行实体识别。在公开数据集CMeEE、CCKS2019、CCKS2020上进行对比实验，实验结果表明，MANM模型在3个数据集上的F1值分别达到64.29%、86.12%、90.32%，验证了所提方法在医疗实体识别中的有效性。

关键词: 命名实体识别, 医疗文本, 注意力机制, 长短时记忆网络, 语义特征

Abstract: Medical entity recognition aims to identify multiple medical entities，such as diseases，symptoms，and drugs，from medical texts. It can support the development of downstream tasks，such as knowledge graphs and smart medical treatment，with high theoretical and practical application value.To address the problems in which the current Named Entity Recognition（NER） model extracts relatively simple semantic features and cannot comprehend the semantics of medical texts，this study proposes a neural network model based on multiple attention mechanism，called MANM.To capture richer semantic features in the texts，prior knowledge of medical vocabulary is first introduced in the model input，and the global semantic features of the medical text are obtained through the self-attention mechanism.The implicit semantic features at the vocabulary and character levels are obtained through the bilinear attention mechanism to determine the feature vectors containing dependencies between characters and words.To improve the contextual information capture ability of the model，the timing sequence features of the texts are obtained through an improved Long and Short-Term Memory（LSTM） network，and a multi-head self-attention mechanism is designed to obtain the implicit associated semantic features between words.Finally，these multi-level semantic features are fused to perform entity recognition using a Conditional Random Field（CRF）.This study conducts a comparative experiment based on public datasets CMeEE，CCKS2019，and CCKS2020.The experimental results show that the F1-score of the three datasets reach 64.29%，86.12%，and 90.32%，respectively，which verifies the effectiveness of the proposed method in medical entity recognition.

Key words: Named Entity Recognition（NER）, medical text, attention mechanism, Long and Short-Term Memory（LSTM） network, semantic feature

中图分类号:

TP18

陈明, 刘蓉, 张晔. 基于多重注意力机制的中文医疗实体识别[J]. 计算机工程, 2023, 49(6): 314-320.

CHEN Ming, LIU Rong, ZHANG Ye. Chinese Medical Entity Recognition Based on Multiple Attention Mechanism[J]. Computer Engineering, 2023, 49(6): 314-320.

https://www.ecice06.com/CN/Y2023/V49/I6/314

图/表 12

20230615171150

20230615171153

20230615171157

20230615171200

20230615171203

20230615171206

20230615171209

20230615171213

20230615171216

20230615171219

20230615171223

20230615171226

参考文献

[1] LI J,SUN A X,HAN J L,et al.A survey on deep learning for named entity recognition[J].IEEE Transactions on Knowledge and Data Engineering,2022,34(1):50-70.
[2] DAI Z H,YANG Z L,YANG Y M,et al.Transformer-XL:attentive language models beyond a fixed-length context[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics.Stroudsburg,USA:Association for Computational Linguistics,2019:2978-2989.
[3] LIN Y,LIU L Y,JI H,et al.Reliability-aware dynamic feature composition for name tagging[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics.Stroudsburg,USA:Association for Computational Linguistics,2019:165-174.
[4] RONRAN C,LEE S.Effect of character and word features in bidirectional LSTM-CRF for NER[C]//Proceedings of IEEE International Conference on Big Data and Smart Computing.Washington D.C.,USA:IEEE Press,2020:613-616.
[5] JIE Z M,LU W.Dependency-guided LSTM-CRF for named entity recognition[EB/OL].[2022-04-01].https://arxiv.org/abs/1909.10148.
[6] YAN C,SU Q,WANG J.MoGCN:mixture of gated convolutional neural network for named entity recognition of Chinese historical texts[J].IEEE Access,2020,8:181629-181639.
[7] LEE L H,LU Y.Multiple embeddings enhanced multi-graph neural networks for Chinese healthcare named entity recognition[J].IEEE Journal of Biomedical and Health Informatics,2021,25(7):2801-2810.
[8] ZHAI F F,POTDAR S,XIANG B,et al.Neural models for sequence chunking[J].Proceedings of the AAAI Conference on Artificial Intelligence,2017,31(1):3365-3371.
[9] SABOUR S,FROSST N,HINTON G E.Dynamic routing between capsules[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems.New York,USA:ACM Press,2017:3859-3869.
[10] SHEN Y Y,YUN H,LIPTON Z,et al.Deep active learning for named entity recognition[C]//Proceedings of the 2nd Workshop on Representation Learning for NLP.Stroudsburg,USA:Association for Computational Linguistics,2017:252-256.
[11] BROWN T B,MANN B,RYDER N,et al.Language models are few-shot learners[C]//Proceedings of the 34th International Conference on Neural Information Processing Systems.New York,USA:ACM Press,2020:1877-1901.
[12] ZHANG D,WEI S Z,LI S S,et al.Multi-modal graph fusion for named entity recognition with targeted visual guidance[C]//Proceedings of the AAAI Conference on Artificial Intelligence.Palo Alto,USA:AAAI,2021:14347-14355.
[13] QIU J H,WANG Q,ZHOU Y M,et al.Fast and accurate recognition of Chinese clinical named entities with residual dilated convolutions[C]//Proceedings of IEEE International Conference on Bioinformatics and Biomedicine.Washington D.C.,USA:IEEE Press,2019:935-942.
[14] 晏阳天,赵新宇,吴贤.基于BERT与字形字音特征的医疗命名实体识别[EB/OL].[2022-03-01].https://bj.bcebos.com/v1/conference/ccks2020/eval_paper/ccks2020_eval_paper_3_1_2.pdf.YAN Y T,ZHAO X Y,WU X.Medical named entity recognition based on BERT and glyph and phonetic features[EB/OL].[2022-03-01].https://bj.bcebos.com/v1/conference/ccks2020/eval_paper/ccks2020_eval_paper_3_1_2.pdf.(in Chinese)
[15] 罗凌,杨志豪,宋雅文,等.基于笔画ELMo和多任务学习的中文电子病历命名实体识别研究[J].计算机学报,2020,43(10):1943-1957.LUO L,YANG Z H,SONG Y W,et al.Chinese clinical named entity recognition based on stroke ELMo and multi-task learning[J].Chinese Journal of Computers,2020,43(10):1943-1957.(in Chinese)
[16] ZHOU B H,CAI X R,ZHANG Y,et al.MTAAL:multi-task adversarial active learning for medical named entity recognition and normalization[C]//Proceedings of the AAAI Conference on Artificial Intelligence.Palo Alto,USA:AAAI,2021:14586-14593.
[17] SONG Y W,YANG Z H,LUO L,et al.Biomedical mutation entity recognition method based on character convolution neural network[J].Journal of Chinese Information Processing,2021,35(5):63-69.
[18] LI X N,YAN H,QIU X P,et al.FLAT:Chinese NER using flat-lattice transformer[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.Stroudsburg,USA:Association for Computational Linguistics,2020:1-10.
[19] DEVLIN J,CHANG M W,LEE K,et al.BERT:pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of NAACL-HLT.Minneapolis,USA:[s.n.],2019:4171-4186.
[20] CUI Y M,CHE W X,LIU T,et al.Pre-training with whole word masking for Chinese BERT[J].IEEE/ACM Transactions on Audio,Speech,and Language Processing,2021,29:3504-3514.
[21] CUI Y,CHE W,LIU T,et al.Revisiting pre-trained models for Chinese natural language processing[C]//Proceedings of EMNLP2020.[S.l.]:SIGDAT,2020:657-668.
[22] LAN Z Z,CHEN M D,GOODMAN S,et al.ALBERT:a lite BERT for self-supervised learning of language representations[EB/OL].[2022-04-01].https://arxiv.org/abs/1909.11942.
[23] LI N,LUO L,DING Z,et al.DUTIR at the CCKS-2019Task1:improving Chinese clinical named entity recognition using stroke ELMo and transfer learning[C]//Proceedings of CCKS2019.Hangzhou,China:Springer,2019:24-27.
[24] LIU M,ZHOU X,CAO Z,et al.Team MSIIP at CCKS 2019 Task 1[C]//Proceedings of CCKS2019.Hangzhou,China:Springer,2019:1-11.
[25] 杨文明,毕金良,邹佳丽,等.基于ChiEHRBert与多模型融合的医疗命名实体识别[EB/OL].[2022-03-01]. https://bj.bcebos.com/v1/conference/ccks2020/eval_paper/ccks2020_eval_paper_3_1_3.pdf.YANG W M,BI J L,ZOU J L,et al.Medical named entity recognition based on ChiEHRBert and multi-model fusion[EB/OL].[2022-03-01].https://bj.bcebos.com/v1/conference/ccks2020/eval_paper/ccks2020_eval_paper_3_1_3.pdf.(in Chinese)

选择文件类型/文献管理软件名称

选择包含的内容