计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (8): 157-162.doi: 10.11896/jsjkx.220700161

• 人工智能 • 上一篇    下一篇

增强实体表示的文档级关系抽取方法研究

丁肖摇1, 周刚1,2, 卢记仓1,2, 陈静1,2   

  1. 1 战略支援部队信息工程大学 郑州 450001
    2 数学工程与先进计算国家重点实验室 郑州 450001
  • 收稿日期:2022-07-18 修回日期:2023-05-24 出版日期:2023-08-15 发布日期:2023-08-02
  • 通讯作者: 周刚(gzhougzhou@126.com)
  • 作者简介:(dingxiaoyao2006@126.com)
  • 基金资助:
    河南省自然科学基金(222300420590)

Study on Enhanced Entity Representation for Document-level Relation Extraction

DING Xiaoyao1, ZHOU Gang1,2, LU Jicang1,2, CHEN Jing1,2   

  1. 1 PLA Strategic Support Force Information Engineering University,Zhengzhou 450001,China
    2 State Key Laboratory of Mathematical Engineering and Advanced Computing,Zhengzhou 450001,China
  • Received:2022-07-18 Revised:2023-05-24 Online:2023-08-15 Published:2023-08-02
  • About author:DING Xiaoyao,born in 1990,Ph.D.His main research interests include relation extraction and knowledge graph.
    ZHOU Gang,born in 1974,Ph.D,professor.His main research interests include big data,knowledge graph and data mining.
  • Supported by:
    Natural Science Foundation of Henan Province,China(222300420590).

PDF (PC)

758

摘要： 文档级关系抽取是自然语言处理领域研究的热点和难点问题,基于图的模型是当前文档级关系抽取的主流方法之一,该类方法虽然能有效解决实体节点之间的长距离依赖问题,但其在构造节点时往往未充分考虑句子上下文、文档主题、实体对距离、实体对相似度等额外信息,导致关系抽取的性能较低。针对该问题,提出了基于增强实体表示的文档级关系抽取模型。首先,将原始文档作为输入,构建基础文档图结构;然后,通过图神经网络传播机制聚合邻接点的信息,将与实体关系预测相关的句子上下文、主题信息融入基础文档图的实体节点表示中,从而获得增强的实体节点表示;最后,利用增强后实体节点的图模型对实体关系进行预测。实验结果表明,所提模型在文档级关系抽取任务中的性能优于已有模型,且可解释性更好。

关键词: 文档级, 关系抽取, 实体表示, 图模型

Abstract: Document-level relation extraction is a hot and challenging issue in natural language processing.Graph-based model is one of the mainstream methods of document-level relation extraction.Although this method can effectively solve the long-distance dependency between entity nodes,it often fails to fully consider the additional information such as sentence context,document topic,entity to entity distance and entity to similarity when constructing nodes,resulting in low performance of relationship extraction.A document-level relation extraction model based on enhanced entity representation is proposed to solve this problem.Firstly,the original document is used as input to construct the basic document graph structure.Then,the graph neural network propagation mechanism is used to aggregate the information of adjacent nodes,and the sentence context and topic information related to entity relation prediction is integrated into the entity node representation of the primary document graph,to obtain an enhanced entity node representation.Finally,the graph model of the enhanced entity node is used to predict the entity relationship.Experimental results show that the performance of the proposed model in the document-level relation extraction task is better than that of the existing models,and has better interpretability.

Key words: ocument-level, Relation extraction, Entity representation, Graph-based model

中图分类号: 

  • TN929.5
[1]SPEER R,CHIN J,HAVASI C.Conceptnet 5.5:An open multilingual graph of general knowledge[C]//Proceedings of the 31st AAAI Conference on Artificial Intelligence.2017:4444-4451.
[2]YU M,YIN W,HASAN K S,et al.Improved neural relation detection for knowledge base question answering[C]//Proceedings of the 55th Annual Meeting of the Association for Computa-tional Linguistics.2017:571-581.
[3]KADRY A,DIETZ L.Open relation extraction for support passage retrieval:merit and open issues[C]//Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval.2017:1149-1152.
[4]MIWA M,BANSAL M.End-to-end relation extraction usinglstms on sequences and tree structures[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics.2016:1105-1116.
[5]ZHANG Y,QI P,MANNING C D.Graph convolution overpruned dependency trees improves relation extraction[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing.2018:2205-2215.
[6]GUO Z,ZHANG Y,LU W.Attention guided graph convolutional networks for relation extraction[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics.2019:241-251.
[7]YAO Y,YE D,LI P,et al.DocRED:Alarge-scale document-level relation extraction dataset [C]//Proceedings of the 57th Conference of the Association for Computational Linguistics.2019:764-777.
[8]SCHLICHTKRULL M,KIPF T N,BLOEM P,et al.Modeling relational data with graph convolutional networks[C]//Euro-pean Semantic Web Conference.2018:593-607.
[9]CHRISTOPOULOU F,MIWA M,ANANIADOU S.Connec-ting the dots:Document-level neural relation extraction with edge-oriented graphs[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing.2019:4924-4935.
[10]NAN G,GUO Z,SEKULIC I,et al.Reasoning with latent structure refinement for document-level realtion extraction[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.2020:1546-1557.
[11]KIPF T,WELLING M.Semi-supervised classification withgraph convolutional networks[C]//Proceedings of the 5th International Conference on Learning Representations.2017.
[12]ZENG S,WU Y,CHANG B.Sire:Separate intra-and inter-sentential reasoning for document-level relation extraction[C]//Findings of the Association for Computational Linguistics.2021:524-534.
[13]XU W,CHEN K,ZHAO T.Document-level relation extraction with reconstruction [C]//Proceedings of the 33th Conference on Artificial Intelligence.2020:14167-14175.
[14]DEVLIN J,CHANG M,LEE K,et al.BERT:Pre-training ofdeep bidirectional transformers for language understanding[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics.2019:4171-4186.
[15]TANG H,CAO Y,ZHANG Z,et al.HIN:Hierarchical infe-rence network for document-level relation extraction[C]//Proceedings of the 24th Pacific-Asia Conference.2020:197-209.
[16]ZENG S,XU R,CHANG B,et al.Double graph based reasoning for document-level relation extraction[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing.2020:1630-1640.
[17]JIA R,WONG C,POON H.Document-level n-ary relation extraction with multiscale representation Learning[C]//Procee-dings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics.2019:3693-3704.
[18]TAO Q,LUO X,WANG H,et al.Enhancing relation extraction using syntactic indicators and sentential contexts[C]//2019 IEEE 31st International Conference on Tools with Artificial intelligence.2019:1574-1580.
[19]HIRANO T,ASANO H,MATSUO Y,et al.Recognizing relation expression between named entities based on inherent and context-dependent features of relational words[C]//Proceedings of the 23th International Conference on Computational Linguistics.2010:409-417.
[20]WU T,KONG F.Document-level relation extraction based ongraph attention convolutional neural network[J].Journal of Chinese Information Processing,2021,35(10):73-80.
[21]YANG C N,PENG D L.Document-level entity relation extraction method integrating bidirectional simple recurrent unit and capsule network [J].Journal of Chinese Computer Systems,2022,43(5):964-968.
[22]YUAN C,HUANG H,FENG C,et al.Document-level relation extraction with entity-selection attention[J].Information Sciences,2021,568:163-174.
[23]WANG D,HU W,CAO E,et al.Global-to-local neural networks for document-level relation Extraction[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing.2020:3711-3721.
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