NAHGNN: Neighborhood Aware Heterogeneous Graph Neural Network

LI Qiang; ZHENG Wei; CHEN Ming; TAN Xing-yi; MA Hua

doi:10.12263/DZXB.20250420

您当前的位置：

首页 >

文章列表页 >

NAHGNN: Neighborhood Aware Heterogeneous Graph Neural Network

PAPERS | 更新时间：2026-02-10

- NAHGNN: Neighborhood Aware Heterogeneous Graph Neural Network
- ACTA ELECTRONICA SINICA Vol. 53, Issue 11, Pages: 4142-4156(2025)
- 作者机构：
  
  湖南师范大学信息科学与工程学院，湖南长沙 410081
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China(62477009);Natural Science Foundation of Hunan Province(2023JJ30411);Major Science and Technology Innovation Platform Project of Hunan Province(2024JC1003);Key Research and Development Program of Hunan Province(2021GK5014)
- DOI：10.12263/DZXB.20250420
  CLC： TP391
- Received：26 May 2025，
  
  Accepted：03 November 2025，
  
  Published：25 November 2025
- 稿件说明：
移动端阅览
李强, 郑唯, 陈明, 等. NAHGNN：邻域感知异构图神经网络[J]. 电子学报, 2025, 53(11): 4142-4156.

LI Qiang, ZHENG Wei, CHEN Ming, et al. NAHGNN: Neighborhood Aware Heterogeneous Graph Neural Network[J]. Acta Electronica Sinica, 2025, 53(11): 4142-4156.
李强, 郑唯, 陈明, 等. NAHGNN：邻域感知异构图神经网络[J]. 电子学报, 2025, 53(11): 4142-4156. DOI：10.12263/DZXB.20250420

LI Qiang, ZHENG Wei, CHEN Ming, et al. NAHGNN: Neighborhood Aware Heterogeneous Graph Neural Network[J]. Acta Electronica Sinica, 2025, 53(11): 4142-4156. DOI：10.12263/DZXB.20250420

摘要

异构图广泛存在于社交网络、推荐系统和生物网络等复杂场景中.基于元路径的异构图神经网络通过定义高阶语义路径对跨类型间接关系进行显式建模，以提升复杂关系建模能力.但现有研究或未加区分地使用指定长度内所有的元路径特征，随着元路径长度的增加，所生成特征的数量呈指数上升，造成语义信息的冗余；或受限于高阶聚合导致的过平滑现象，造成边缘信息的丢失.为解决这些问题，本文提出了一种异构图神经网络模型（Neighborhood Aware Heterogeneous Graph Neural Network，NAHGNN），从邻域感知的角度切入，通过任务解耦，将特征生成分为两个步骤：关联元路径生成和邻域感知特征聚合.首先，关联元路径生成模块利用起始节点与结束节点均是目标节点类型的关联元路径特征，学习目标节点间丰富的语义信息.其次，从目标节点的邻域感知方式出发，设计了一个简单高效的邻域感知特征聚合模块，对关联元路径中忽略的邻域信息进行提取.最后，为了拟合相应邻域感知方式的语义表示，避免邻域感知特征间相互影响，设计了一个带掩码的语义融合模块，融合不同特征间的语义信息.在DBLP、ACM、IMDB和Freebase四个公开异构图数据集上与六种主流异构图神经网络基线进行实验对比.结果表明，NAHGNN在节点分类任务中Micro-F1提升幅度为0.63~12.50个百分点，训练时间与GPU内存消耗显著下降，并展现出良好的可解释性.

Abstract

Heterogeneous graphs are widely present in complex scenarios such as social networks

recommendation systems

and biological networks. Meta-path-based heterogeneous graph neural networks (HGNNs) explicitly model cross-type indirect relationships via high-order semantic paths

enhancing the ability to capture complex dependencies. However

existing studies either use all meta-path features within a specified length without distinction

leading to redundancy in semantic information as the number of generated features rises exponentially with the increase of the meta-paths length

or suffer from over-smoothing caused by high-order aggregation

resulting in the loss of edge information. To address these issues

this paper proposes a neighborhood aware heterogeneous graph neural network (NAHGNN). From the perspective of neighborhood awareness and through task decoupling

the feature generation is divided into two steps: associative meta-path generation and neighborhood-aware feature aggregation. Firstly

an associative meta-path generation module learns rich semantic information between target nodes by leveraging associative meta-path features that both start and end nodes are of the target type. Secondly

a simple and efficient neighborhood-aware feature aggregation module is designed based on the neighborhood-aware modalities of target nodes to extract neglected neighborhood information in associative meta-paths. Finally

to fit the semantic representations of corresponding neighborhood-aware modalities and avoid mutual interference between neighborhood-aware features

a semantic fusion module with a band mask is designed to integrate semantic information across different features. Experimental comparisons are conducted with six mainstream heterogeneous graph neural network baselines on four public heterogeneous graph datasets (DBLP

ACM

IMDB

and Freebase). The results show that NAHGNN achieves a Micro-F1 improvement of 0.63 to 12.50 percentage points in node classification tasks

significantly reduces training time and GPU memory consumption

and exhibits favorable interpretability.

关键词

Keywords

references

KHEMANI B , PATIL S , KOTECHA K , et al . A review of graph neural networks: Concepts, architectures, techniques, challenges, datasets, applications, and future directions [J ] . Journal of Big Data , 2024 , 11 ( 1 ): 18 .

ANAND V , MAURYA A K . A survey on recommender systems using graph neural network [J ] . ACM Transactions on Information Systems , 2025 , 43 ( 1 ): 1 - 49 .

CHEN M , LI H K , PAN Y H , et al . Multi-filter based signed heterogeneous graph convolutional networks for predicting activating/inhibiting drug-target interactions [J ] . Methods , 2025 , 241 : 51 - 58 .

AI W , WEI Y Y , SHAO H E , et al . Edge-enhanced minimum-margin graph attention network for short text classification [J ] . Expert Systems with Applications , 2024 , 251 : 124069 .

FU X Y , ZHANG J N , MENG Z Q , et al . MAGNN: Metapath aggregated graph neural network for heterogeneous graph embedding [C ] // Proceedings of The Web Conference 2020 . New York : ACM , 2020 : 2331 - 2341 .

MAVROMATIS C , KARYPIS G . GNN-RAG: Graph neural retrieval for efficient large language model reasoning on knowledge graphs [C ] // Findings of the Association for Computational Linguistics: ACL 2025 . Stroudsburg : ACL , 2025 : 16682 - 16699 .

SU X R , HU P W , YOU Z H , et al . Dual-channel learning framework for drug-drug interaction prediction via relation-aware heterogeneous graph transformer [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2024 , 38 ( 1 ): 249 - 256 .

CHANG Y M , CHEN C , HU W B , et al . Megnn: Meta-path extracted graph neural network for heterogeneous graph representation learning [J ] . Knowledge-Based Systems , 2022 , 235 : 107611 .

CHEN M , JIANG Y J , LEI X J , et al . Drug-target interactions prediction based on signed heterogeneous graph neural networks [J ] . Chinese Journal of Electronics , 2024 , 33 ( 1 ): 231 - 244 .

LIU J , SONG L Y , WANG G T , et al . Meta-HGT: Metapath-aware HyperGraph Transformer for heterogeneous information network embedding [J ] . Neural Networks , 2023 , 157 : 65 - 76 .

范伟 , 周魏 , 文俊浩 . 基于异构图的双通道交叉自适应对比学习推荐 [J ] . 电子学报 , 2023 , 51 ( 7 ): 1929 - 1938 .

FAN W , ZHOU W , WEN J H . Recommendation based on graph heterogeneous using dual channel cross-adaptive contrast learning [J ] . Acta Electronica Sinica , 2023 , 51 ( 7 ): 1929 - 1938 . (in Chinese)

WANG B H , CHEN J W , LI C D , et al . Distributionally robust graph-based recommendation system [C ] // Proceedings of the ACM Web Conference 2024 . New York : ACM , 2024 : 3777 - 3788 .

HU Z N , DONG Y X , WANG K S , et al . Heterogeneous graph transformer [C ] // Proceedings of The Web Conference 2020 . New York : ACM , 2020 : 2704 - 2710 .

FU X Y , KING I . MECCH: Metapath context convolution-based heterogeneous graph neural networks [J ] . Neural Networks , 2024 , 170 : 266 - 275 .

MAO Q H , LIU Z M , LIU C H , et al . HINormer: Representation learning on heterogeneous information networks with graph transformer [C ] // Proceedings of the ACM Web Conference 2023 . New York : ACM , 2023 : 599 - 610 .

SCHLICHTKRULL M , KIPF T N , BLOEM P , et al . Modeling relational data with graph convolutional networks [C ] // The Semantic Web . Cham : Springer , 2018 : 593 - 607 .

YANG X C , YAN M Y , PAN S R , et al . Simple and efficient heterogeneous graph neural network [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2023 , 37 ( 9 ): 10816 - 10824 .

KIPF T N , WELLING M . Semi-supervised classification with graph convolutional networks [EB/OL ] .( 2017-02-22 )[ 2025-10-10 ] . https://arXiv.org/abs/1609.02907 https://arXiv.org/abs/1609.02907 .

HAMILTON W L , YING R , LESKOVEC J . Inductive representation learning on large graphs [C ] // Proceedingsofthe31st International Conference on Neural Information Processing Systems . New York : ACM , 2017 : 1025 - 1035 .

VELIČKOVIĆ P , CUCURULL G , CASANOVA A , et al . Graph attention networks [EB/OL ] . ( 2018-02-04 )[ 2025-10-09 ] . https://arxiv.org/abs/1710.10903 https://arxiv.org/abs/1710.10903 .

ZHU S C , ZHOU C , PAN S R , et al . Relation structure-aware heterogeneous graph neural network [C ] // 2019 IEEE International Conference on Data Mining . Piscataway : IEEE , 2020 : 1534 - 1539 .

LV Q S , DING M , LIU Q , et al . Are we really making much progress?:Revisiting, benchmarking and refining heterogeneous graph neural networks [C ] // Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining . New York : ACM , 2021 : 1150 - 1160 .

BUTLER L , PARADA-MAYORGA A , RIBEIRO A . Convolutional learning on multigraphs [J ] . IEEE Transactions on Signal Processing , 2023 , 71 : 933 - 946 .

AHN H , YANG Y Y , GAN Q , et al . Descent steps of a relation-aware energy produce heterogeneous graph neural networks [EB/OL ] . ( 2022-10-21 )[ 2025-09-30 ] . https://arXiv.org/abs/2206.11081 https://arXiv.org/abs/2206.11081 .

HE L C , BAI L , YANG X , et al . High-order graph attention network [J ] . Information Sciences , 2023 , 630 : 222 - 234 .

ZHANG C X , SONG D J , HUANG C , et al . Heterogeneous graph neural network [C ] // Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining . New York : ACM , 2019 : 793 - 803 .

WANG X , JI H Y , SHI C , et al . Heterogeneous graph attention network [C ] // The World Wide Web Conference . New York : ACM , 2019 : 2022 - 2032 .

LI C , YAN Y Y , FU J H , et al . HetReGAT-FC: Heterogeneous residual graph attention network via feature completion [J ] . Information Sciences , 2023 , 632 : 424 - 438 .

LI C , FU J H , YAN Y Y , et al . Higher order heterogeneous graph neural network based on node attribute enhancement [J ] . Expert Systems with Applications , 2024 , 238 : 122404 .

VAN D M L , HINTON G . Visualizing data using t-SNE [J ] . Journal of Machine Learning Research , 2008 , 9 ( 11 ): 2579 - 2605 .

ZHANG Z J , WANG X , ZHOU H C , et al . Can large language models improve the adversarial robustness of graph neural networks? [C ] // Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1 . New York : ACM , 2025 : 2008 - 2019 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Sequence-Based and Cross-Modal Alignment Model for Protein Function Prediction

Neighborhood and Hypergraph Collaboration for Session-Based Recommendation

Object Detection Based on EIMYOLO for High-Resolution Remote Sensing Images

Single-Image High Dynamic Range Reconstruction Based on Multi-Attention and Perceptual Weighted Learning

RAMAN SCATTERING IN GaN GROWN BY VPE

Related Author

XU Min

HU Chun-ling

HU Ting

ZHANG Fang-fang

DAI Xiang-long

CHEN Rong-yuan

WEN Jie-bin

HUANG Shao-nian

Related Institution

School of Artificial Intelligence and Big Data, Hefei University

College of Frontier Intersection, Hunan University of Technology and Business

Key Laboratory of Hunan Province for Statistical Learning and Intelligent Computation, Hunan University of Technology and Business

School of Computer Science, Hunan University of Technology and Business

School of Information and Technology, Shanxi University

⁰