| 31 | 0 | 17 |
| 下载次数 | 被引频次 | 阅读次数 |
针对大规模社交网络中节点关系复杂、社区结构隐蔽导致社区发现算法精度较低的问题,提出一种基于图注意力机制和多距离分析的社区发现算法。首先,在预训练部分引入多层感知机模块提高样本的识别精度。其次,提出基于图注意力机制的读出操作模块,通过自适应地学习节点在图表示学习中的贡献权重,实现加权聚合操作,从而增强图表示的判别能力,提升社区发现精度和鲁棒性。最后,设计一种多距离分析方法,增强对小目标的感知能力,提升检测准确率。实验结果表明,改进模型在多个数据集上均表现优异。Amazon数据集上的F1值达到了90.22%,相比CLARE和ProCom模型分别提高了11.33个百分点和5.86个百分点。Jaccard相似度则达到了83.46%,相比CLARE和ProCom模型分别提高了14.96个百分点和7.62个百分点。
Abstract：To address the low accuracy of community detection in large-scale social networks, which was caused by complex node relationships and hidden community structures, a community detection algorithm based on graph attention mechanisms and multi-distance analysis was proposed. Firstly, a multilayer perceptron(MLP) module was introduced during the pre-training phase so that the accuracy of sample recognition could be improved. Secondly, a graph attention-based readout module was proposed, by which the contribution weights of nodes in graph representation learning were adaptively learned. The weighted aggregation was enabled, the discriminative power of graph-level representations was enhanced, and the accuracy and robustness of community detection were improved. Finally, a multi-distance analysis method was designed to strengthen the detection of small communities and to further increase overall detection accuracy. It was demonstrated by experimental results that superior performance was delivered by the improved model across multiple datasets, On the Amazon dataset, the F1 score of 90. 22% was achieved, which represented improvements of 11. 33 and 5. 86 percentage points over the CLARE and ProCom models, respectively. For Jaccard similarity, a score of 83. 46% was obtained, corresponding to improvements of 14. 96 and 7. 62 percentage points over CLARE and ProCom, respectively.
[1]吴日铭,韩益亮,郭凯阳,等.推荐算法背景下兴趣社区挖掘算法[J].郑州大学学报(理学版), 2023,55(6):77-83.WU R M, HAN Y L, GUO K Y, et al. Interest community discovery algorithm in recommendation network[J].Journal of Zhengzhou university(natural science edition), 2023, 55(6):77-83.
[2]李悦,谢珺,侯文丽,等.融合用户偏好优化聚类的协同过滤推荐算法[J].郑州大学学报(理学版),2020,52(2):29-35.LI Y, XIE J, HOU W L, et al. Collaborative filtering recommendation algorithm based on optimized clustering with user preference[J]. Journal of Zhengzhou university(natural science edition),2020,52(2):29-35.
[3]刘振鹏,董姝慧,李泽园,等.面向社交网络数据的等差数列聚类匿名算法[J].郑州大学学报(理学版), 2022, 54(1):41-47.LIU Z P, DONG S H, LI Z Y, et al. Anonymity algorithm of arithmetic sequence clustering for social network data[J]. Journal of Zhengzhou university(natural science edition), 2022, 54(1):41-47.
[4]YANG J, LESKOVEC J. Overlapping community detection at scale:a nonnegative matrix factorization approach[C]∥Proceedings of the 6th ACM International Conference on Web Search and Data Mining. New York:ACM Press, 2013:587-596.
[5]CAVALLARI S, ZHENG V, CAI H Y, et al. Learning community embedding with community detection and node embedding on graphs[C]∥Proceedings of the 26th ACM International Conference on Information and Knowledge Management. New York:ACM Press, 2017:377-386.
[6]JIA Y T, ZHANG Q Q, ZHANG W N, et al. CommunityGAN:community detection with generative adversarial nets[C]∥Proceedings of the 28th International World Wide Web Conference. New York:ACM Press, 2019:784-794.
[7]BAKSHI A, PARTHASARATHY S, SRINIVASAN K.Semi-supervised community detection using structure and size[C]∥2018 IEEE International Conference on Data Mining. Piscataway:IEEE Press, 2018:869-874.
[8]ZHANG Y, XIONG Y, YE Y, et al. SEAL:learning heuristics for community detection with generative adversarial networks[C]∥Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery&Data Mining. New York:ACM Press, 2020:1406-1416.
[9]WU X X, XIONG Y, ZHAN Y, et al. CLARE:a semisupervised community detection algorithm[C]∥Proceedings of the 28th ACM SIGKDD International Conference on Knowledge Discovery&Data Mining. New York:ACM Press, 2022:1903-1913.
[10]WU X X, XIONG K Y, XIONG Y, et al. ProCom:a few-shot targeted community detection algorithm[C]∥Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York:ACM Press, 2024:1-11.
[11]YINGR,YOUJX,MORRISC,etal.Hierarchical graphrepresentationlearningwithdifferentiablepooling/[EB OL].(2018-07-22)[2025-05-30].https:∥arxiv.//org pdf1806. 08804.
[12]LESTER B, AL-RFOU R, CONSTANT N. The power of scale for parameter-efficient prompt tuning[C]∥Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. Stroudsburg:ACL Press,2021:3045-3059.
[13]LIU P F, YUAN W Z, FU J L, et al. Pre-train, prompt,and predict:a systematic survey of prompting methods in natural language processing[J]. ACM computing surveys, 2023, 55(9):1-35.
[14]HAN K, WANG Y, CHEN H, et al. Vision MLP:hierarchical vision MLP with recursive gated aggregation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(9):10502-10519.
[15]SUN Y D, ZHU D J, WANG Y S, et al. GTC:GNNTransformer co-contrastive learning for self-supervised heterogeneous graph representation[J]. Neural networks,2025, 181:106645.
[16]ZHANG X Y, ZHANG Y K, CHEN A L, et al. Optimizing multi label student performance prediction with GNNTINet:a contextual multidimensional deep learning framework[J]. PLoS one, 2025, 20(1):e0314823.
[17]HAKKAL S, AIT LAHCEN A. Leveraging graph neural network for learner performance prediction[J]. Expert systems with applications, 2025, 293:128724.
[18]FU C Z, SU Y K, SU K F, et al. HAM-GNN:a hierarchical attention-based multi-dimensional edge graph neural network for dialogue act classification[J]. Expert systems with applications, 2025, 261:125459.
[19]LI G H, MULLER M, THABET A, et al. DeepGCNs:can GCNs go as deep as CNNs?[C]∥2019 IEEE/CVF International Conference on Computer Vision. Piscataway:IEEE Press, 2021:9267-9276.
[20]GAO D L, QIN H L, LI F H. Research on the comprehensive performance evaluation method of genetic algorithm based on euclidean distance minimization[C]∥2024 8th International Workshop on Control Engineering and Advanced Algorithms. Piscataway:IEEE Press,2025:76-80.
[21]NGUYEN D D N, SOOD K, NOSOUHI M R, et al. RF fingerprinting-based IoT node authentication using mahalanobis distance correlation theory[J]. IEEE networking letters, 2022, 4(2):78-81.
[22]丁志祥.基于MLP神经网络的尿液指标分析模型[J].建模与仿真, 2024(5):5329-5336.DING Z X. Urine parameter analysis model based on MLP neural network[J]. Modeling and simulation, 2024(5):5329-5336.
基本信息:
DOI：10.13705/j.issn.1671-6841.2025085
中图分类号:O157.5;TP18
引用信息:
[1]张震,吴国豪,张红霞,等.基于图注意力机制和多距离分析的社区发现算法[J].郑州大学学报(理学版)().DOI:10.13705/j.issn.1671-6841.2025085.
基金信息:
河南省重点研发专项项目(231111211600); 河南省重大公益专项项目(201300311200)
2026-04-24
2026-04-24
2026-04-24