计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (1): 160-169.doi: 10.11896/jsjkx.231100117

• 数据库&大数据&数据科学 • 上一篇    下一篇

路径掩码自编码器引导无监督属性图节点聚类

丁新宇1, 孔兵1, 陈红梅1, 包崇明2, 周丽华1   

  1. 1 云南大学信息学院 昆明 650504
    2 云南大学软件学院 昆明 650504
  • 收稿日期:2023-11-19 修回日期:2024-04-27 出版日期:2025-01-15 发布日期:2025-01-09
  • 通讯作者: 孔兵(kongbing@ynu.edu.cn)
  • 作者简介:(13623489952@163.com)
  • 基金资助:
    国家自然科学基金(62062066,61762090,61966036,62276227);云南省基础科研项目(202201AS070015);云南省中青年学术和技术带头人后备人才项目(202205AC160033);云南省智能系统与计算重点实验室(202205AG070003);云南大学专业学位研究生实践创新项目(ZC-23234311)

Path-masked Autoencoder Guiding Unsupervised Attribute Graph Node Clustering

DING Xinyu1, KONG Bing1, CHEN Hongmei1, BAO Chongming2, ZHOU Lihua1   

  1. 1 School of Information Science and Engineering,Yunnan University,Kunming 650504,China
    2 School of Software,Yunnan University,Kunming 650504,China
  • Received:2023-11-19 Revised:2024-04-27 Online:2025-01-15 Published:2025-01-09
  • About author:DING Xinyu,born in 1997,postgra-duate,is a member of CCF(No.R3213G).His main research interests include deep graph clustering and data mining.
    KONG Bing,born in 1968,Ph.D,asso-ciate professor.His main research in-terests include social network analysis and machine learning.
  • Supported by:
    National Natural Science Foundation of China(62062066,61762090,61966036,62276227),Yunnan Fundamental Research Projects(202201AS070015),Young and Middle-aged Academic and Technical Leaders Reserve Talent Project in Yunnan Province(202205AC160033),Yunnan Key Laboratory of Intelligent Systems and Computing(202205AG070003) and Practical Innovation Project of Postgraduate Students in the Professional Degree of Yunnan University(ZC-23234311).

PDF (PC)

170

摘要： 图聚类的目的在于发现网络的社区结构。针对目前聚类方法无法很好地获取网络深层潜在社区信息,且不能对特征进行合适的信息整合导致节点社区语义不清晰的问题,提出了一种路径掩码自编码器引导无监督属性图节点聚类模型(Path-Masked Autoencoder Guiding Unsupervised Attribute Graph Node Clustering,PAUGC)。该模型通过对网络进行随机路径掩码后使用自编码器来深度挖掘网络拓扑结构,从而获得良好的全局结构语义信息,利用规范性方法来对特征进行信息整合,使节点特征能够更好地表征特征的类别信息。此外,模型结合模块最大化来抓取整个图中的底层社区群落信息,目的在于更合理地将其融合到低维度节点特征中。最后通过自训练聚类来不断迭代优化更新聚类表示以获得最终的节点特征。通过在8个基准数据集上与11种经典方法进行大量实验对比,证明了PAUGC的有效性。

关键词: 深度图聚类, 无监督学习, 特征信息整合, 模块最大化, 聚类自训练

Abstract: The purpose of graph clustering is to discover the community structure of the network.Aiming at the problem that the current clustering methods can not well obtain the deep potential community information of the network,and can not make sui-table information integration of the features,resulting in unclear semantics of the node community,a path-masked autoencoder guiding unsupervised attribute graph node clustering(PAUGC)model is proposed.This model utilizes an autoencoder to deeply dig the network topology structure by randomly masking network paths,thereby obtaining excellent global structural semantic information.Utilizing a normative method for information integration of the features,so that the node features are able to better characterize the class information of the features.In addition,the model combines modularity maximization to capture the under-lying community clusters information in the whole graph,aiming to more reasonably fuse it into the low-dimensional node features.Finally,the model iteratively optimizes and updates the clustering representation through self-training clustering to obtain the final node features.By conducting extensive experiments and comparisons with 11 classical methods on 8 benchmark datasets,PAUGC has been proven to be effective compared to current mainstream methods.

Key words: Deep graph clustering, Unsupervised learning, Feature integration, Module maximization, Self-training for clustering

中图分类号: 

  • TP391
[1]LIU C,WEN L,KANG Z,et al.Self Supervised Consensus Rep-resentation Learning for Attributed Graph[C]//Proceedings of the 29th ACM International Conference on Multimedia.New York:ACM Press,2021:2654-2662.
[2]WANG M,WANG C,YU J X,et al.Community Detection inSocial Networks:An In Depth Benchmarking Study with A Procedure Oriented Framework[J].Proceedings of the VLDB Endowment,2015,8(10):998-1009.
[3]GARCIA J O,ASHOURVAN A,MULDOON S,et al.Applications of Community Detection Techniques to Brain Graphs:Algorithmic Considerations and Implications for Neural Function[J].Proceedings of the IEEE,2018,106(5):846-867.
[4]KRISHNAMURTHY B,WANG J.On Network-Aware Clustering of Web Clients[C]//Proceedings of the Conference on Applications,Technologies,Architectures,and Protocols for Computer Communication.New York:ACM Press,2000:97-110.
[5]CHIANG M M T,MIRKIN B.Intelligent Choice of the Number of Clusters in K-Means Clustering:An Experimental Study with Different Cluster Spreads[J].Journal of Classification,2010,27:3-40.
[6]MCINNES L,HEALY J,ASTELS S.HDBSCAN:HierarchicalDensity Based Clustering[J].Open Source Softw,2017,2(11):205.
[7]KIPF T N,WELLING M.Semi-Supervised ClassificAtion with Graph Convolutional Networks[J].arXiv:1609.02907,2016.
[8]TANG J M,HAN H,HUANG L.Coarse grained and Fine-grained Features Extraction Based on Unsupervised Learning in Pedestrian Reidentification[J].Computer Engineering,2022,48(4):269-275,283.
[9]WANG C,PAN S,LONG G,et al.Mgae:Marginalized GraphAutoencoder for Graph Clustering[C]//Proceedings of the 2017 ACM on Conference on Information and Knowledge Management.Singapore:ACM Press,2017:889-898.
[10]PARK J,LEE M,CHANG H J,et al.Symmetric Graph Convolutional Autoencoder for Unsupervised Graph Representation Learning[C]//Proceedings of the IEEE International Confe-rence on Computer Vision.Seoul:IEEE Press,2019:6519-6528.
[11]PAN S,HU R,LONG G,et al.Adversarially Regularized Graph Autoencoder for Graph Embedding[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Stockholm:IJCAI Press,2018:2609-2615.
[12]WANG C,PAN S,HU R,et al.Attributed Graph Clustering:A Deep Attentional Embedding Approach[C]//Proceedings of the 28th International Joint Conference on Artificial Intelligence.Macao:IJCA Press,2019:3670-3676.
[13]SALEHI A,DAVULCU H.Graph Attention Auto-Encoders[C]//Proceedings of the 32nd IEEE International Conference on Tools with Artificial Intelligence.ELECTR NETWORK:IEEE Press,2020:989-996.
[14]HUI B,ZHU P,HU Q.Collaborative Graph Convolutional Networks:Unsupervised Learning Meets Semi-Supervised Learning[C]//Proceedings of the AAAI Conference on Artificial Intelligence.New York:AAAI Press,2020:4215-4222.
[15]ZHOU X,SU L,LI X,et al.Community Detection Based on Unsupervised Attributed Network Embedding[J].Expert Systems with Applications,2023,213:118937.
[16]NEWMAN M E J.Modularity and Community Structure in Networks[J].Proceedings of the National Academy of Sciences,2006,103(23):8577-8582.
[17]YANG X,LIU Y,ZHOU S,et al.Cluster Guided Contrastive Graph Clustering Network[C]//Proceedings of the 37th AAAI Conference on Artificial Intelligence.Washington:AAAI Press,2023:10834-10842.
[18]ZHANG S,PANG J,LIAO M X.Graph Clustering Algorithm Based on Hybrid Feature Selection[J].Journal of Chinese Computer Systems,2024,45(3):606-612.
[19]SCHAUB M T,DELVENNE J C,ROSVALL M,et al.TheMany Facets of Community Detection in Complex Networks[J].Applied Network Science,2017,2(1):1-13.
[20]FORTUNATO S.Community Detection in Graphs[J].Physics Reports,2010,486(3/4/5):75-174.
[21]FORTUNATO S,HRIC D.Community Detection in Networks:A User Guide[J].Physics Reports,2016,100(659):1-44.
[22]AGARWAL G,KEMPE D.Modularity Maximizing Graph Communities via Mathematical Programming[J].The European Physical Journal B,2008,66(3):409-418.
[23]KIPF T N,WELLING M.Variational Graph Auto-Encoders[J].arXiv:1611.07308,2016.
[24]QIU C,HUANG Z,XU W,et al.VGAER:Graph Neural Network Reconstruction Based Community Detection[J].arXiv:2201.04066,2022.
[25]LI J,YU J,LI J,et al.Dirichlet Graph Variational Autoencoder[J].Advances in Neural Information Processing Systems,2020,33:5274-5283.
[26]TIAN Y,DONG K,ZHANG C,et al.Heterogeneous GraphMasked Autoencoders[C]//Proceedings of the 37th AAAI Conference on Artificial Intelligence.Washington:AAAI Press,2023:9997-10005.
[27]XU H,XIA W,GAO Q,et al.Graph Embedding Clustering:Graph Attention Auto Encoder with Cluster-Specificity Distribution[J].Neural Networks,2021,142:221-230.
[28]CUI G,ZHOU J,YANG C,et al.Adaptive Graph Encoder for Attributed Graph Embedding[C]//Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Disco-very & Data Mining.Electr Network:KDD Press,2020:976985.
[29]PENG Z,LIU H,JIA Y,et al.Deep Attention Guided Graph Clustering with Dual Self Supervision[J].IEEE Transactions on Circuits and Systems for Video Technology,2023,33(7):3296-3307.
[30]BO D,WANG X,SHI C,et al.Structural Deep Clustering Network[C]//Proceedings of the 29th World Wide Web Confe-rence.2020:1400-1410.
[31]TU W,ZHOU S,LIU X,et al.Deep Fusion Clustering Network[C]//Proceedings of the AAAI Conference on Artificial Intelligence.Electr Network:AAAI Press,2021:9978-9987.
[32]HASSANI K,KHASAHMADI A H.Contrastive Multi-ViewRepresentation Learning on Graphs[C]//Proceedings of the 37th International Conference on Machine Learning.Electr Network:ICML,2020:4074-4084.
[33]GONG L,ZHOU S,TU W,et al.Attributed Graph Clustering with Dual Redundancy Reduction[C]//Proceedings of the 31st International Joint Conference on Artificial Intelligence.Vienna:IJCAI Press,2022:3015-3021.
[34]LEE N,LEE J,PARK C.Augmentation-Free Self-SupervisedLearning on Graphs[C]//Proceedings of the 36th AAAI Conference on Artificial Intelligence.Electr Network:AAAI Press,2022:7372-7380.
[35]ZHAO H,YANG X,WANG Z,et al.Graph Debiased Contrastive Learning with Joint Representation Clustering[C]//Proceedings of the 30th International Jo int Conference on Artificial Intelligence.Electr Network:IJCAI Press,2021:3434-3440.
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