HMSG: Heterogeneous graph neural network based on Metapath SubGraph learning

doi:10.1016/j.knosys.2023.110930

CORC > 自动化研究所 > 中国科学院自动化研究所 > 复杂系统认知与决策实验室

	HMSG: Heterogeneous graph neural network based on Metapath SubGraph learning
	Guan, Mengya 1,5; Cai, Xinjun 1,5; Shang, Jiaxing 1,5; Hao, Fei 2; Liu, Dajiang 1,5; Jiao, Xianlong 1,5; Ni, Wancheng 3,4
刊名	KNOWLEDGE-BASED SYSTEMS
	2023-11-04
卷号	279 页码:11
关键词	Graph neural network Heterogeneous graph Metapath Subgraph Information network
ISSN号	0950-7051
DOI	10.1016/j.knosys.2023.110930
通讯作者	Shang, Jiaxing(shangjx@cqu.edu.cn)
英文摘要	Heterogeneous graph neural network (HGNN) models, capable of learning low-dimensional dense vectors from heterogeneous graphs for downstream graph-mining tasks, have attracted increasing attention in recent years. For these models, metapath-based methods have been widely adopted. However, most existing metapath-based HGNN models either discard intermediate nodes within a metapath, resulting in information loss, or indiscriminately aggregate information along a metapath containing different types of nodes, resulting in unavoidable learning bias. To overcome these limitations, a new HGNN model named HMSG, is proposed in this paper to comprehensively capture structural, semantic and attribute information from both homogeneous and heterogeneous neighbors more purposefully. To achieve this, a type-specific linear transformation is first applied to transfer the node attributes to different types of nodes with the same latent factor space. In the new model, the heterogeneous graph is decomposed into multiple metapath-based homogeneous and heterogeneous subgraphs where each subgraph associates specific semantic and structural information; this is different from existing models, which mainly rely on symmetric metapaths. Subsequently, tailored attention-based message aggregation methods are independently applied to each subgraph such that information learning can be more targeted. Finally, information from different subgraphs is fused through graph-level attention to obtain a complete representation. The learned representations are evaluated by several graph-mining tasks. Results indicate that the HMSG attains the best performance in all evaluation metrics than state-of-the-art baselines. Further ablation experiments demonstrate the effectiveness of the modules designed for the HMSG. (c) 2023 Elsevier B.V. All rights reserved.
资助项目	National Key R&D Program of China[2022YFC3006400] ; National Key R&D Program of China[2022YFC3006402] ; National Natural Science Foundation of China[61966008] ; National Natural Science Foundation of China[U2033213] ; National Natural Science Foundation of China[62072064] ; Natural Science Foundation of Chongqing, China[CSTB2022NSCQ-MSX1017] ; Open Fund of Key Laboratory of Dependable Service Computing in Cyber Physical Society, China[CPSDSC202207] ; Open Fund of the Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences, China[CASIA-KFKT-10] ; Fundation, China[2020-JCJQ-ZD- 087-00]
WOS关键词	MAXIMIZATION
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:001076369600001
资助机构	National Key R&D Program of China ; National Natural Science Foundation of China ; Natural Science Foundation of Chongqing, China ; Open Fund of Key Laboratory of Dependable Service Computing in Cyber Physical Society, China ; Open Fund of the Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences, China ; Fundation, China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/53092]
专题	复杂系统认知与决策实验室
通讯作者	Shang, Jiaxing
作者单位	1.Chongqing Univ, Key Lab Dependable Serv Comp Cyber Phys Soc, Minist Educ, Chongqing, Peoples R China 2.Shaanxi Normal Univ, Sch Comp Sci, Xian, Peoples R China 3.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China 4.Univ Chinese Acad Sci, Beijing, Peoples R China 5.Chongqing Univ, Coll Comp Sci, Chongqing, Peoples R China
推荐引用方式 GB/T 7714	Guan, Mengya,Cai, Xinjun,Shang, Jiaxing,et al. HMSG: Heterogeneous graph neural network based on Metapath SubGraph learning[J]. KNOWLEDGE-BASED SYSTEMS,2023,279:11.
APA	Guan, Mengya.,Cai, Xinjun.,Shang, Jiaxing.,Hao, Fei.,Liu, Dajiang.,...&Ni, Wancheng.(2023).HMSG: Heterogeneous graph neural network based on Metapath SubGraph learning.KNOWLEDGE-BASED SYSTEMS,279,11.
MLA	Guan, Mengya,et al."HMSG: Heterogeneous graph neural network based on Metapath SubGraph learning".KNOWLEDGE-BASED SYSTEMS 279(2023):11.