Heterogeneous graph attention network for relation extraction based on feature combination

doi:10.11772/j.issn.1001-9081.2024081076

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (8): 2470-2476.DOI: 10.11772/j.issn.1001-9081.2024081076

• The 21th CCF Conference on Web Information Systems and Applications (WISA 2024) • Previous Articles

Heterogeneous graph attention network for relation extraction based on feature combination

Jiaxin YAN¹^,²^,³, Yanping CHEN¹^,²^,³(), Weizhe YANG¹^,²^,³, Ruizhang HUANG¹^,²^,³, Yongbin QIN¹^,²^,³

^1.Engineering Research Center of Ministry of Education for Text Computing and Cognitive Intelligence，Guizhou University，Guiyang Guizhou 550025，China
^2.State Key Laboratory of Public Big Data （Guizhou University），Guiyang Guizhou 550025，China
^3.College of Computer Science and Technology，Guizhou University，Guiyang Guizhou 550025，China

Received:2024-08-01 Revised:2024-08-12 Accepted:2024-08-15 Online:2024-09-12 Published:2025-08-10
Contact: Yanping CHEN
About author:YAN Jiaxin， born in 2000， M. S. candidate. His research interests include natural language processing， information extraction.
YANG Weizhe， born in 1996， Ph. D. candidate. His research interests include natural language processing.
HUANG Ruizhang， born in 1979， Ph. D.， professor. Her research interests include data fusion analysis， text mining， network mining， knowledge discovery.
QIN Yongbin， born in 1980， Ph. D.， professor. His research interests include big data management and application， multi-source data fusion.
Supported by:
Major Science and Technology Foundation of Guizhou Province(［2024］003);National Key Research and Development Program of China(2023YFC3304500);National Natural Science Foundation of China(62166007)

基于特征组合的异构图注意力网络关系抽取

闫家鑫¹^,²^,³, 陈艳平¹^,²^,³(), 杨卫哲¹^,²^,³, 黄瑞章¹^,²^,³, 秦永彬¹^,²^,³

^1.贵州大学文本计算与认知智能教育部工程研究中心，贵阳 550025
^2.公共大数据国家重点实验室（贵州大学），贵阳 550025
^3.贵州大学计算机科学与技术学院，贵阳 550025

通讯作者: 陈艳平
作者简介:闫家鑫（2000—），男，四川成都人，硕士研究生，CCF会员，主要研究方向：自然语言处理、信息抽取
杨卫哲（1996—），男，河南新乡人，博士研究生，主要研究方向：自然语言处理
黄瑞章（1979—），女，天津人，教授，博士，CCF会员，主要研究方向：数据融合分析、文本挖掘、网络挖掘、知识发现
秦永彬（1980—），男，山东烟台人，教授，博士，CCF高级会员，主要研究方向：大数据管理与应用、多源数据融合。
基金资助:
贵州省科学技术基金重点资助项目(［2024］003);国家重点研发计划项目(2023YFC3304500);国家自然科学基金资助项目(62166007)

Abstract

Abstract:

Relation extraction aims to identify predefined semantic relationships between two entities within a sentence. Traditional graph neural network-based relation extraction methods generally use dependency trees to construct a graphical representation structure of the sentence. However， the graph structure constructed by dependency tree has limited expression ability and is unable to fully capture rich syntactic structure information of the target entity. To address these issues， a relation extraction method of Heterogeneous Graph ATtention network （HGAT） based on feature combination was proposed. Firstly， atomic features were extracted from the sentence， and composite features were obtained by combining these atomic features. Secondly， the composite features and relation labels were represented as two types of nodes on the heterogeneous graph to construct a “feature-relation bipartite graph”. Finally， a graph attention network was used to update the nodes dynamically to perform relation extraction. In this method， the composite features and the syntactic structure information in the sentence were utilized effectively， thereby enhancing the performance of relation extraction. Experimental results on ACE05 English dataset and SemEval-2010 task 8 dataset show that this method achieves F1-scores of 84.11% and 90.67%， respectively， demonstrating the effectiveness of the proposed method.

Key words: relation extraction, atomic feature, feature combination, heterogeneous graph, graph attention network

摘要：

关系抽取旨在从句子中提取2个实体之间的预定义语义关系。传统基于图神经网络的关系抽取方法一般通过依赖树构建句子的图表示结构；然而，使用依赖树构建出的图结构表达能力单一，且无法完整捕捉到目标实体丰富的语法结构信息。针对这些问题，提出基于特征组合的异构图注意力网络（HGAT）关系抽取方法。首先，抽取句子中的原子特征，并通过组合这些原子特征得到句子的组合特征；其次，把组合特征和关系标签表示为异构图上的两种节点以构建“特征-关系二部图”；最后，使用图注意力网络动态地更新节点，进而实现关系抽取。所提方法能有效利用组合特征和句子中的语法结构信息，进而提升关系抽取的性能。在ACE05英文数据集和SemEval-2010 task 8数据集上的实验结果表明，所提方法分别达到了84.11%和90.67%的F1值，证明了所提方法的有效性。

关键词: 关系抽取, 原子特征, 特征组合, 异构图, 图注意力网络

CLC Number:

TP391

Jiaxin YAN, Yanping CHEN, Weizhe YANG, Ruizhang HUANG, Yongbin QIN. Heterogeneous graph attention network for relation extraction based on feature combination[J]. Journal of Computer Applications, 2025, 45(8): 2470-2476.

闫家鑫, 陈艳平, 杨卫哲, 黄瑞章, 秦永彬. 基于特征组合的异构图注意力网络关系抽取[J]. 《计算机应用》唯一官方网站, 2025, 45(8): 2470-2476.

Figures/Tables 9

References 28

[1]	ZHANG Y， ZHONG V， CHEN D， et al. Position-aware attention and supervised data improve slot filling［C］// Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Stroudsburg： ACL， 2017： 35-45.
[2]	文坤建，陈艳平，黄瑞章，等. 基于提示学习的生物医学关系抽取方法［J］. 计算机科学， 2023， 50（10）： 223-229.
	WEN K J， CHEN Y P， HUANG R Z， et al. Biomedical relationship extraction method based on prompt learning［J］. Computer Science， 2023， 50（10）： 223-229.
[3]	XIA R， DING Z. Emotion-cause pair extraction： a new task to emotion analysis in texts［C］// Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Stroudsburg： ACL， 2019： 1003-1012.
[4]	ZHANG Y， QI P， MANNING C D. Graph convolution over pruned dependency trees improves relation extraction［C］// Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. Stroudsburg： ACL， 2018： 2205-2215.
[5]	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. Stroudsburg： ACL， 2019： 241-251.
[6]	CHEN Y， ZHENG Q， CHEN P. A set space model for feature calculus［J］. IEEE Intelligent Systems， 2017， 32（5）： 36-42.
[7]	CHEN Y， WANG G， ZHENG Q， et al. A set space model to capture structural information of a sentence［J］. IEEE Access， 2019， 7： 142515-142530.
[8]	CHEN Y， ZHENG Q， CHEN P. Feature assembly method for extracting relations in Chinese［J］. Artificial Intelligence， 2015， 228： 179-194.
[9]	CHEN Y， YANG W， WANG K， et al. A neuralized feature engineering method for entity relation extraction［J］. Neural Networks， 2021， 141： 249-260.
[10]	WANG H， CHEN Y， YANG W， et al. A two dimensional feature engineering method for relation extraction［EB/OL］. ［2024-08-10］..
[11]	KHAN I， KWON Y W. Multi-class malware detection via deep graph convolutional networks using TF-IDF-based attributed call graphs［C］// Proceedings of the 2023 International Conference on Information Security Applications， LNCS 14402. Singapore： Springer， 2024： 188-200.
[12]	杨卫哲，秦永彬，黄瑞章，等. 面向中文关系抽取的句子结构获取方法［J］. 数据采集与处理， 2021， 36（3）： 605-620.
	YANG W Z， QIN Y B， HUANG R Z， et al. Sentence structure acquisition method for Chinese relation extraction［J］. Journal of Data Acquisition and Processing， 2021， 36（3）：605-620.
[13]	HU Y， CHEN Y， HUANG R， et al. A hierarchical convolutional model for biomedical relation extraction［J］. Information Processing and Management， 2024， 61（1）： No.103560.
[14]	衡红军，徐天宝. 基于多尺度卷积和门控机制的注意力情感分析模型［J］. 计算机应用， 2022， 42（9）： 2674-2679.
	HENG H J， XU T B. Attention sentiment analysis model based on multi-scale convolution and gating mechanism［J］. Journal of Computer Applications， 2022， 42（9）： 2674-2679.
[15]	NGUYEN T H， GRISHMAN R. Combining neural net-works and log-linear models to improve relation extraction［EB/OL］. ［2024-08-10］..
[16]	ZHENG S， XU J， ZHOU P， et al. A neural network frame-work for relation extraction： learning entity semantic and relation pattern［J］. Knowledge-Based Systems， 2016， 114： 12-23.
[17]	ZHANG S， ZHENG D， HU X， et al. Bidirectional long short-term memory networks for relation classification［C］// Proceedings of the 29th Pacific Asia Conference on Language， Information and Computation. Stroudsburg： ACL， 2015： 73-78.
[18]	TIAN Y， CHEN G， SONG Y， et al. Dependency-driven relation extraction with attentive graph convolutional networks［C］// Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing （Volume 1： Long Papers）. Stroudsburg： ACL， 2021： 4458-4471.
[19]	ZHAO K， XU H， CHENG Y， et al. Representation iterative fusion based on heterogeneous graph neural network for joint entity and relation extraction［J］. Knowledge-Based Systems， 2021， 219： No.106888.
[20]	CHEN Y， ZHENG Q， ZHANG W. Omni-word feature and soft constraint for Chinese relation extraction［C］// Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics （Volume 1： Long Papers）. Stroudsburg： ACL， 2014： 572-581.
[21]	GORMLEY M R， YU M， DREDZE M. Improved relation extraction with feature-rich compositional embedding models［C］// Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. Stroudsburg： ACL， 2015： 1774-1784.
[22]	CHRISTOPOULOU F， MIWA M， ANANIADOU S. A walk-based model on entity graphs for relation extraction［C］// Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics （Volume 2： Short Papers）. Stroudsburg： ACL， 2018： 81-88.
[23]	WU Y， CHEN Y， QIN Y， et al. A recollect-tuning method for entity and relation extraction［J］. Expert Systems with Applications， 2024， 245： No.123000.
[24]	SUN C， GONG Y， WU Y， et al. Joint type inference on entities and relations via graph convolutional networks［C］// Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Stroudsburg： ACL， 2019： 1361-1370.
[25]	XU J， CHEN Y， QIN Y， et al. A feature combination-based graph convolutional neural network model for relation extraction［J］. Symmetry， 2021， 13（8）： No.1458.
[26]	XU J， CHEN Y， QIN Y， et al. A learnable graph convolutional neural network model for relation extraction［C］// Proceedings of the 28th China Conference on Information Retrieval， LNCS 13819. Cham： Springer， 2023： 90-104.
[27]	CHEN G， TIAN Y， SONG Y， et al. Relation extraction with type-aware map memories of word dependencies［C］// Findings of the Association for Computational Linguistics： ACL-IJCNLP 2021. Stroudsburg： ACL， 2021： 2501-2512.
[28]	唐媛，陈艳平，扈应，等. 基于多尺度混合注意力卷积神经网络的关系抽取模型［J］. 计算机应用， 2024， 44（7）： 2011-2017.
	TANG Y， CHEN Y P， HU Y， et al. Relation extraction based on multi-scale mixed attention convolutional neural networks［J］. Journal of Computer Applications， 2024， 44（7）： 2011-2017.

原子特征	函数表示	描述
f₁	｛f = LeftPosOf（e_i ）｝	实体i左近邻单词词性
f₂	｛f = RightPosOf（e_i ）｝	实体i右近邻单词词性
f₃	｛f = TypeOf（e_i ）｝	实体i的类型
f₄	｛f = SubTypeOf（e_i ）｝	实体i的子类型
f₅	｛f = HeadOf（e_i ）｝	实体i的中心名词
f₆	｛f = PositionBetween（e₁，e₂）｝	两实体相对位置关系

原子特征	函数表示	描述
f₁	｛f = LeftPosOf（e_i ）｝	实体i左近邻单词词性
f₂	｛f = RightPosOf（e_i ）｝	实体i右近邻单词词性
f₃	｛f = TypeOf（e_i ）｝	实体i的类型
f₄	｛f = SubTypeOf（e_i ）｝	实体i的子类型
f₅	｛f = HeadOf（e_i ）｝	实体i的中心名词
f₆	｛f = PositionBetween（e₁，e₂）｝	两实体相对位置关系

特征组合	函数表示
F₁	｛F = LeftPosOf（e₁）⊕TypeOf（e₁）｝
F₂	｛F = TypeOf（e₁）⊕RightPosOf（e₁）｝
F₃	｛F = LeftPosOf（e₂）⊕TypeOf（e₂）｝
F₄	｛F = TypeOf（e₂）⊕RightPosOf（e₂）｝
F₅	｛F = TypeOf（e₁）⊕TypeOf（e₂）｝
F₆	｛F = SubTypeOf（e₁）⊕SubTypeOf（e₂）｝
F₇	｛F = HeadOf（e₁）⊕HeadOf（e₂）｝

特征组合	函数表示
F₁	｛F = LeftPosOf（e₁）⊕TypeOf（e₁）｝
F₂	｛F = TypeOf（e₁）⊕RightPosOf（e₁）｝
F₃	｛F = LeftPosOf（e₂）⊕TypeOf（e₂）｝
F₄	｛F = TypeOf（e₂）⊕RightPosOf（e₂）｝
F₅	｛F = TypeOf（e₁）⊕TypeOf（e₂）｝
F₆	｛F = SubTypeOf（e₁）⊕SubTypeOf（e₂）｝
F₇	｛F = HeadOf（e₁）⊕HeadOf（e₂）｝

数据集	样本数			关系数
数据集	训练集	验证集	测试集	关系数
ACE05	83 293	10 412	10 412	7
SemEval-2010 task 8	8 000	—	2 717	19

Heterogeneous graph attention network for relation extraction based on feature combination

基于特征组合的异构图注意力网络关系抽取

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 9

References 28

Related Articles 15

Recommended Articles

Metrics

模型	精确率	召回率	F1值
FCM^［21］	71.52	49.32	58.26
Walk-based^［22］	69.70	59.50	64.20
Re-tuning ^［23］	75.90	71.40	73.60
GCN^［24］	68.70	65.40	67.00
FC-GCN^［25］	85.64	75.95	80.50
L-GCN^［26］	87.09	79.79	83.28
A-GCN^［18］	—	—	79.05
本文模型	89.39	79.73	84.11

模型	精确率	召回率	F1值
C-GCN^［4］	—	—	84.80
TaMM^［27］	—	—	90.06
Mix Attention^［28］	91.69	89.06	90.32
A-GCN^［18］	—	—	89.85
本文模型	91.46	89.91	90.67

模型	ACE05	SemEval-2010 task 8
本文模型	84.11	90.67
-Combined features	83.08	90.06
-Cf.Syntactic features	83.36	90.38
-Cf.Structural features	83.50	90.40
-HGAT	82.13	89.98
-Gate mechanism	83.78	90.43

[1]	Shuo ZHANG, Guokai SUN, Yuan ZHUANG, Xiaoyu FENG, Jingzhi WANG. Dynamic detection method of eclipse attacks for blockchain node analysis [J]. Journal of Computer Applications, 2025, 45(8): 2428-2436.
[2]	Haijie WANG, Guangxin ZHANG, Hai SHI, Shu CHEN. Document-level relation extraction based on entity representation enhancement [J]. Journal of Computer Applications, 2025, 45(6): 1809-1816.
[3]	Wenjing YAN, Ruidong WANG, Min ZUO, Qingchuan ZHANG. Recipe recommendation model based on hierarchical learning of flavor embedding heterogeneous graph [J]. Journal of Computer Applications, 2025, 45(6): 1869-1878.
[4]	Dawei YANG, Xihai XU, Wei SONG. Relation extraction method combining semantic enhancement and perception attention [J]. Journal of Computer Applications, 2025, 45(6): 1801-1808.
[5]	Jie HU, Cui WU, Jun SUN, Yan ZHANG. Document-level relation extraction model based on anaphora and logical reasoning [J]. Journal of Computer Applications, 2025, 45(5): 1496-1503.
[6]	Jiaxin LI, Site MO. Power work order classification in substation area based on MiniRBT-LSTM-GAT and label smoothing [J]. Journal of Computer Applications, 2025, 45(4): 1356-1362.
[7]	Bin LI, Min LIN, Siriguleng, Yingjie GAO, Yurong WANG, Shujun ZHANG. Joint entity-relation extraction method for ancient Chinese books based on prompt learning and global pointer network [J]. Journal of Computer Applications, 2025, 45(1): 75-81.
[8]	Jianpeng HU, Lichen ZHANG. Deep spatio-temporal network model for multi-time step wind power prediction [J]. Journal of Computer Applications, 2025, 45(1): 98-105.
[9]	Liang ZHU, Jingzhe MU, Hongqiang ZUO, Jingzhong GU, Fubao ZHU. Location privacy-preserving recommendation scheme based on federated graph neural network [J]. Journal of Computer Applications, 2025, 45(1): 136-143.
[10]	Xianglan WU, Yang XIAO, Mengying LIU, Mingming LIU. Text-to-SQL model based on semantic enhanced schema linking [J]. Journal of Computer Applications, 2024, 44(9): 2689-2695.
[11]	Hang YANG, Wanggen LI, Gensheng ZHANG, Zhige WANG, Xin KAI. Multi-layer information interactive fusion algorithm based on graph neural network for session-based recommendation [J]. Journal of Computer Applications, 2024, 44(9): 2719-2725.
[12]	Fan YANG, Yao ZOU, Mingzhi ZHU, Zhenwei MA, Dawei CHENG, Changjun JIANG. Credit card fraud detection model based on graph attention Transformation neural network [J]. Journal of Computer Applications, 2024, 44(8): 2634-2642.
[13]	Yubo ZHAO, Liping ZHANG, Sheng YAN, Min HOU, Mao GAO. Relation extraction between discipline knowledge entities based on improved piecewise convolutional neural network and knowledge distillation [J]. Journal of Computer Applications, 2024, 44(8): 2421-2429.
[14]	Yuan TANG, Yanping CHEN, Ying HU, Ruizhang HUANG, Yongbin QIN. Relation extraction model based on multi-scale hybrid attention convolutional neural networks [J]. Journal of Computer Applications, 2024, 44(7): 2011-2017.
[15]	Dianhui MAO, Xuebo LI, Junling LIU, Denghui ZHANG, Wenjing YAN. Chinese entity and relation extraction model based on parallel heterogeneous graph and sequential attention mechanism [J]. Journal of Computer Applications, 2024, 44(7): 2018-2025.