计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (10): 337-343.doi: 10.11896/jsjkx.230800033

• 人工智能 • 上一篇    下一篇

基于多关系视图轴向注意力的文档级关系抽取

吴皓1, 周刚1,2, 卢记仓1,2, 刘洪波1,2, 陈静1,2   

  1. 1 战略支援部队信息工程大学数据与目标工程学院 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:2023-08-07 修回日期:2023-12-13 出版日期:2024-10-15 发布日期:2024-10-11
  • 通讯作者: 周刚(gzhougzhou@126.com)
  • 作者简介:(wuhao186000@163.com)
  • 基金资助:
    河南省科技攻关项目(222102210081)

Document-level Relation Extraction Based on Multi-relation View Axial Attention

WU Hao1, ZHOU Gang1,2, LU Jicang1,2, LIU Hongbo1,2, CHEN Jing1,2   

  1. 1 School of Data and Target Engineering,Information Engineering University,Zhengzhou 450001,China
    2 State Key Laboratory of Mathematical Engineering and Advanced Computing,Zhengzhou 450001,China
  • Received:2023-08-07 Revised:2023-12-13 Online:2024-10-15 Published:2024-10-11
  • About author:WU Hao,born in 1997,postgraduate.His main research interests include relation extraction and data mining.
    ZHOU Gang,born in 1974,Ph.D,professor.His main research interests include knowledge graph,mass data processing and knowledge grapgh.
  • Supported by:
    Science and Technology Research Program of the Department of Science and Technology of Henan Province(222102210081).

PDF (PC)

251

摘要： 文档级关系抽取旨在从文档中提取多个实体之间的关系。针对现有工作在不同关系类型的条件下,对于实体间的多跳推理能力受限的问题,提出了一种基于多关系视图轴向注意力的文档级关系抽取模型。该模型将依据实体间的关系类型构建多视图的邻接矩阵,并基于该多视图的邻接矩阵进行多跳推理。基于两个文档级关系抽取基准数据集GDA和DocRED分别进行实验,结果表明,所提模型在生物数据集GDA上的F1指标达到85.7%,性能明显优于基线模型;在DocRED数据集上也能够有效捕获实体间的多跳关系。

关键词: 关系抽取, 文档级, 轴向注意力, 多视图, 多跳推理

Abstract: Document-level relationship extraction aims to extract relationships between multiple entities from documents.To address the limited multi-hop reasoning capacity of existing methods for establishing connections between entities with different relationship types,this paper propose a document-level relationship extraction model based on multi-relation view axial attention.The model will construct a multi-view adjacency matrix based on the relationship types between entities,and use it to perform multi-hop reasoning.In order to evaluate the proposed model's performance,two benchmark datasets for document-level relationship extraction,namely GDA and DocRED are used in this study.The experimental results demonstrate that the F1 metric achieves 85.7% on the biological dataset GDA,significantly surpassing the baseline model's performance.Moreover,the proposed model proves effective in capturing the multi-hop relationships among entities in the DocRED dataset.

Key words: Relation extraction, Document level, Axial attention, Multi view, Multi-hop inference

中图分类号: 

  • TP391
[1]WANG Y,YU B,ZHANG Y,et al.TPLinker:Single-stage Joint Extraction of Entities and Relations Through Token Pair Lin-king[J].arXiv:2010.13415,2020.
[2]YE D,LIN Y,DU J,et al.Coreferential Reasoning Learning for Language Representation[C]//Proceedings of the 2020 Confe-rence on Empirical Methods in Natural Language Processing(EMNLP).Association for Computational Linguistics,2020:7170-7186.
[3]ZHANG N,DENG S,SUN Z,et al.Relation Adversarial Network for Low Resource Knowledge Graph Completion[C]//Proceedings of The Web Conference 2020.Taipei,Taiwan:ACM,2020:1-12.
[4]WU T,LI X,LI Y F,et al.Curriculum-Meta Learning for Order-Robust Continual Relation Extraction[C]//Proceedings of the AAAI Conference on Artificial Intelligenc.2021.
[5]YU H,ZHANG N,DENG S,et al.Bridging Text and Know-ledge with Multi-Prototype Embedding for Few-Shot Relational Triple Extraction[J].arXiv:2010.16059,2020.
[6]ZHENG H,WEN R,CHEN X,et al.PRGC:Potential Relation and Global Correspondence Based Joint Relational Triple Extraction[J].arXiv:2106.09895,2021.
[7]YAO Y,YE D,LI P,et al.DocRED:A Large-Scale Document-Level Relation Extraction Dataset[J].arXiv:1906.06127,2019.
[8]VASWANI A,SHAZEER N,PARMAR N,et al.Attention Is All You Need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems.2017:6000-6010.
[9]ZHOU W,HUANG K,MA T,et al.Document-Level Relation Extraction with Adaptive Thresholding and Localized ContextPooling[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2021:14612-14620.
[10]ZHANG N,CHEN X,XIE X,et al.Document-level RelationExtraction as Semantic Segmentation.Proceedings of the Thir-tieth International Joint Conference on Artificial Intelligence[C]//International Joint Conferences on Artificial Intelligence Organization.2021:3999-4006.
[11]LI J,XU K,LI F,et al.MRN:A Locally and Globally Mention-Based Reasoning Network for Document-Level Relation Extraction[C]//Findings of the Association for Computational Linguistics:ACL-IJCNLP 2021.Association for Computational Linguistics,2021:1359-1370.
[12]ZHANG L,CHENG Y.A Masked Image Reconstruction Net-work for Document-level Relation Extraction[J].arXiv:2204.09851,2022.
[13]TAN Q,HE R,BING L,et al.Document-Level Relation Extraction with Adaptive Focal Loss and Knowledge Distillation[J].arXiv:2203.10900,2022.
[14]WU Y,LOU R,LEUNG H C M,et al.RENET:A Deep Lear-ning Approach for Extracting Gene-Disease Associations from Literature[C]//International Conference on Research in Computational Molecular Biology.2019:272-284.
[15]DEVLIN J,CHANG M W,LEE K,et al.BERT:Pre-training of Deep Bidirectional Transformers for Language Understanding[C]//Proceedings of naacL-HLT.2019.
[16]MICIKEVICIUS P,NARANG S,ALBEN J,et al.Mixed Preci-sion Training[J].arXiv.1710.03740,2017.
[17]LOSHCHILOV I,HUTTER F.Decoupled Weight Decay Regularization[J].arXiv:1711.05101,2017.
[18]GOYAL P,DOLLAR P,GIRSHICK R,et al.Accurate,LargeMinibatch SGD:Training ImageNet in 1 Hour[J].arXiv:1706.02677,2017.
[19]WANG H,FOCKE C,SYLVESTER R,et al.Fine-tune Bert for DocRED with Two-step Process[J].arXiv:1909.11898,2019.
[20]XU B,WANG Q,LYU Y,et al.Entity Structure Within and Throughout:Modeling Mention Dependencies for Document-Level Relation Extraction[J]. arXiv.2102.10249,2021.
[21]YU J,YANG D,TIAN S.Relation-Specific Attentions over En-tity Mentions for Enhanced Document-Level Relation Extraction[J].arXiv:2205.14393,2022.
[22]TANG H,CAO Y,ZHANG Z,et al.Hin:Hierarchical inference network for document-level relation extraction[C]//Advances in Knowledge Discovery and Data Mining:24th Pacific-Asia Conference.2020:197-209.
[23]CHRISTOPOULOU F,MIWA M,ANANIADOU S,et al.Con-necting 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.
[24]NAN G,GOU Z,SEKULIC I,et al.Reasoning with LatentStructure Refinement for Document-Level Relation Extraction[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.Association for Computational Linguistics,2020:1546-1557.
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