双重图卷积神经网络驱动的隐藏社区发现算法

doi:10.3778/j.issn.1002-8331.2408-0215

摘要/Abstract

摘要： 隐藏社区检测有助于揭示网络深层次功能和结构特征，是一个具有挑战性的研究领域。隐藏社区由弱关系连接而成，受具有较强连接关系的显性社区影响，在网络中不易被检测到。当前的隐藏社区发现算法对节点属性信息和全局拓扑结构的综合利用仍显不足，为解决这一问题，提出了一种基于双重图卷积神经网络（GCN）联合优化隐藏社区发现算法——HCDGCN（hidden community detection based on dual GCN）。HCDGCN融合节点局部和全局结构特征，通过两个GCN共同迭代优化一个损失函数，并逐步削弱权重，使得弱关系社区变得清晰可见，实现了隐藏社区发现。在真实数据集上的实验结果表明，HCDGCN在隐藏社区发现方面优于现有基准方法，实现了更快的收敛速度和更优的社区划分。

关键词: 社区发现, 隐藏社区发现, 图卷积神经网络, 联合优化

Abstract: Hidden community detection is a challenging research topic that helps to uncover the underlying structural and functional features of networks. Explicit communities with stronger connections impact hidden communities, which are hard to find in networks because they are made up of weak links. The current hidden community discovery techniques do not fully use global topological structure and node attribute information. In order to tackle this problem, it provides HCDGCN (hidden community detection based on dual GCN), a joint optimization technique for hidden community detection based on dual graph convolutional networks (GCNs). By progressively weakening the weights and jointly maximizing a loss function using two GCNs, HCDGCN integrates local and global structural aspects of nodes, achieving hidden community identification and increasing the visibility of weak connection communities. Results from experiments on actual datasets demonstrate that HCDGCN performs better and converges more quickly than current benchmark techniques in hidden community discovery, achieving superior community partitioning.

Key words: community discovery, hidden community discovery, graph convolutional neural network, joint optimization

王小刚, 刘旭. 双重图卷积神经网络驱动的隐藏社区发现算法[J]. 计算机工程与应用, 2025, 61(13): 329-337.

WANG Xiaogang, LIU Xu. Hidden Community Detection Algorithm Driven by Dual Graph Convolutional Networks[J]. Computer Engineering and Applications, 2025, 61(13): 329-337.

参考文献

[1] FORTUNATO S. Community detection in graphs[J]. Physics Reports, 2010, 486(3/4/5): 75-174.
[2] ALBERT R, BARABáSI A L. Statistical mechanics of complex networks[J]. Reviews of Modern Physics, 2002, 74(1): 47-97.
[3] NEWMAN M J. Coauthorship networks and patterns of scientific collaboration[J]. Proceedings of the National Academy of Sciences of the United States of America, 2004, 101(Suppl 1): 5200-5205.
[4] QIAO Y Q, LUO X Y, LI C L, et al. Heterogeneous graph-based joint representation learning for users and POIs in location-based social network[J]. Information Processing & Management, 2020, 57(2): 102151.
[5] NADIMI-SHAHRAKI M H, MOEINI E, TAGHIAN S, et al. DMFO-CD: a discrete moth-flame optimization algorithm for community detection[J]. Algorithms, 2021, 14(11): 314.
[6] LUO Y H, ZHU Z C, WU J S, et al. Exploring habitat patch clusters based on network community detection to identify restored priority areas of ecological networks in urban areas[J]. Urban Forestry & Urban Greening, 2022, 78: 127771.
[7] YANG L, CAO X, HE D, et al. Modularity based community detection with deep learning[C]//Proceedings of the 25th International Joint Conference on Artificial Intelligence, 2016: 2252-2258.
[8] RUSTAMAJI H C, KUSUMA W A, NURDIATI S, et al. Community detection with greedy modularity disassembly strategy[J]. Scientific Reports, 2024, 14(1): 4694.
[9] ANANDKUMAR A, GE R, HSU D, et al. A tensor approach to learning mixed membership community models[J]. The Journal of Machine Learning Research, 2014, 15(1): 2239-2312.
[10] HE D X, LIU D Y, JIN D, et al. A stochastic model for detecting heterogeneous link communities in complex networks[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2015.
[11] 陈可佳, 陈利明, 吴桐. 多层网络社区发现研究综述[J]. 计算机科学与探索, 2020, 14(11): 1801-1812.
CHEN K J, CHEN L M, WU T. Survey on community detection in multi-layer networks[J]. Journal of Frontiers of Computer Science and Technology, 2020, 14(11): 1801-1812.
[12] 朱玉龙, 刘建忠, 张寅宝, 等. 自编码模块化增强非负矩阵分解社区检测算法[J]. 计算机工程与应用, 2024, 60(11): 258-267.
ZHU Y L, LIU J Z, ZHANG Y B, et al. Community detection algorithm with autoencoding-like modular enhanced non-negative matrix factorization[J]. Computer Engineering and Applications, 2024, 60(11): 258-267.
[13] NAIK D, RAMESH D, GANDOMI A H, et al. Parallel and distributed paradigms for community detection in social networks: a methodological review[J]. Expert Systems with Applications, 2022, 187: 115956.
[14] HE K, SOUNDARAJAN S, CAO X Z, et al. Revealing multiple layers of hidden community structure in networks[J]. arXiv:1501.05700, 2015.
[15] KHAWAJA F R, SHENG J F, WANG B, et al. Uncovering hidden community structure in multi-layer networks[J]. Applied Sciences, 2021, 11(6): 2857.
[16] HE K, LI Y R, SOUNDARAJAN S, et al. Hidden community detection in social networks[J]. Information Sciences, 2018, 425: 92-106.
[17] WANG M, LI B Y, HE K, et al. Uncovering the local hidden community structure in social networks[J]. ACM Transactions on Knowledge Discovery from Data, 2023, 17(5): 1-25.
[18] NIU Y Y, KONG D T, LIU L G, et al. Overlapping community detection with adaptive density peaks clustering and iterative partition strategy[J]. Expert Systems with Applications, 2023, 213: 119213.
[19] BERAHMAND K, LI Y F, XU Y. DAC-HPP: deep attributed clustering with high-order proximity preserve[J]. Neural Computing and Applications, 2023, 35(34): 24493-24511.
[20] ZHANG H Q, LU G Q, ZHAN M M, et al. Semi-supervised classification of graph convolutional networks with Laplacian rank constraints[J]. Neural Processing Letters, 2022, 54(4): 2645-2656.
[21] 郑裕龙, 陈启买, 贺超波, 等. 图卷积网络增强的非负矩阵分解社区发现方法[J]. 计算机工程与应用, 2022, 58(11): 73-83.
ZHENG Y L, CHEN Q M, HE C B, et al. Nonnegative matrix factorization community detection method enhanced by graph convolutional network[J]. Computer Engineering and Applications, 2022, 58(11): 73-83.
[22] LANCICHINETTI A, RADICCHI F, RAMASCO J J, et al. Finding statistically significant communities in networks[J]. PLoS One, 2011, 6(4): e18961.
[23] AHN Y Y, BAGROW J P, LEHMANN S. Link communities reveal multiscale complexity in networks[J]. Nature, 2010, 466(7307): 761-764.
[24] FARAGE C, EDLER D, EKL?F A, et al. Identifying flow modules in ecological networks using Infomap[J]. Methods in Ecology and Evolution, 2021, 12(5): 778-786.
[25] BLONDEL V D, GUILLAUME J L, LAMBIOTTE R, et al. Fast unfolding of communities in large networks[J]. Journal of Statistical Mechanics: Theory and Experiment, 2008, 2008(10): 10008.
[26] ZHOU X C, SU L T, LI X J, et al. Community detection based on unsupervised attributed network embedding[J]. Expert Systems with Applications, 2023, 213: 118937.
[27] JANNESARI V, KESHVARI M, BERAHMAND K. A novel nonnegative matrix factorization-based model for attributed graph clustering by incorporating complementary information[J]. Expert Systems with Applications, 2024, 242: 122799.
[28] TRAUD A L, MUCHA P J, PORTER M A. Social structure of facebook networks[J]. Physica A: Statistical Mechanics and Its Applications, 2012, 391(16): 4165-4180.
[29] YANG J, LESKOVEC J. Defining and evaluating network communities based on ground-truth[C]//Proceedings of the ACM SIGKDD Workshop on Mining Data Semantics, 2012: 1-8.