动态视图融合矩阵驱动的多视图矩阵分解聚类算法

doi:10.3778/j.issn.1002-8331.2411-0064

摘要/Abstract

摘要： 多视图聚类能整合不同视图互补信息，揭示数据潜在结构。其中，基于矩阵分解的方法同时优化矩阵分解和聚类时复杂度较高，分开处理则导致聚类依赖固定的视图融合矩阵，会限制对复杂数据结构的表达。为此，提出了动态视图融合矩阵驱动的多视图矩阵分解聚类算法。该算法将矩阵分解与证据C均值聚类整合到一个框架中，并引入超参数平衡两部分重要性。先利用非负矩阵分解获取原始数据低维表示，随后通过智能权重分配机制对各视图融合。融合矩阵与聚类动态更新得到最终结果，这种动态更新策略能够避免依靠固定的融合矩阵实现聚类任务而导致的局限性。在5个真实数据集上验证了该算法的性能，结果证实了其优越性。

关键词: 多视图聚类, 矩阵分解, 智能权重分配机制, C-均值

Abstract: Multi-view clustering can integrate the complementary information of different views and reveal the underlying structure of data. Among them, the method based on matrix factorization has high complexity when optimizing matrix factorization and clustering at the same time, and the clustering depends on fixed view fusion matrix when processing separately, which will limit the expression of complex data structures. Therefore, multi-view matrix factorization clustering algorithm based on dynamic view fusion matrix is proposed. The algorithm integrates matrix factorization and evidence C-means clustering into one framework, and introduces the importance of hyperparameter balance. Firstly, the low-dimensional representation of the original data is obtained by non-negative matrix factorization, and then the views are fused by intelligent weight allocation mechanism. Finally, the fusion matrix and cluster are dynamically updated to get the final result. This dynamic update strategy can avoid the limitations caused by relying on fixed fusion matrix to realize the clustering task. The performance of the proposed algorithm is verified on 5 real data sets, and the results show its superiority.

Key words: multi-view clustering, matrix factorization, intelligent weight allocation mechanism, C-means

李顺勇, 曹利娜, 刘坤, 赵兴旺. 动态视图融合矩阵驱动的多视图矩阵分解聚类算法[J]. 计算机工程与应用, 2025, 61(17): 171-184.

LI Shunyong, CAO Lina, LIU Kun, ZHAO Xingwang. Multi-View Matrix Factorization Clustering Algorithm Based on Dynamic View Fusion Matrix[J]. Computer Engineering and Applications, 2025, 61(17): 171-184.

参考文献

[1] YANG Z B, TAN Y H. The methods for improving large-scale multi-view clustering efficiency: a survey[J]. Artificial Intelligence Review, 2024, 57(6): 1-38.
[2] FU L L, LIN P F, VASILAKOS, et al. An overview of recent multi-view clustering[J]. Neurocomputing, 2020, 402: 148-161.
[3] WANG Y X, LIU X W, QI Y F, et al. A review of multi-view clustering algorithms[C]//Proceedings of the 2023 International Conference on Image Processing, Computer Vision and Machine Learning. New York: IEEE, 2023: 847-851.
[4] CHEN M S, LIN J Q, LI X L, et al. Representation learning in multi-view clustering: a literature review[J]. Data Science and Engineering, 2022, 7(3): 225-241.
[5] HARI S, MUHAMMA D, YUSOF F, et al. Breaking down multi-view clustering: a comprehensive review of multi-view approaches for complex data structures[J]. Engineering Applications of Artificial Intelligence, 2024, 132: 1-21.
[6] KUMAR A, RAI P, DAUME H. Co-regularized multi-view spectral clustering[C]//Advances in Neural Information Processing Systems , 2011: 1413-1421.
[7] WANG R, NIE F P, WANG Z, et al. Parameter-free weighted multi-view projected clustering with structured graph learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2019, 32(10): 2014-2025.
[8] 余瑶, 杜世强, 宋金梅. 面向多视图聚类的低秩张量表示学习[J]. 计算机工程与应用, 2022, 58(13): 154-163.
YU Y, DU S Q, SONG J M. Low-rank tensor representation learning for multi-view clustering[J]. Computer Engineering and Applications, 2022, 58(13): 154-163.
[9] 黄展鹏, 吴杰康, 易法令. 自适应图融合的缺失多视图聚类算法[J]. 计算机工程与应用, 2023, 59(9): 176-181.
HUANG Z P, WU J K, YI F L. Incomplete multi-view clustering algorithm with adaptive graph fusion[J]. Computer Engineering and Applications, 2023, 59(9): 176-181.
[10] ZHANG T, LIU X, GONG L, et al. Late fusion multiple kernel clustering with local kernel alignment maximization[J]. IEEE Transactions on Multimedia, 2023, 25: 993-1007.
[11] ZHANG J P, LI L, WANG S W, et al. Multiple kernel clustering with dual noise minimization[C]//Proceedings of the 30th ACM International Conference on Multimedia. New York: ACM, 2022: 3440-3450.
[12] LI L, WANG S, LIU X, et al. Local sample-weighted multiple kernel clustering with consensus discriminative graph[J]. IEEE Transactions on Neural Networks and Learning Systems, 2024, 35(2): 1721-1734.
[13] 郭圣, 仲兆满, 李存华. 基于深度自编码的多视图子空间聚类网络[J]. 计算机工程与应用, 2020, 56(17): 60-68.
GUO S, ZHONG Z M, LI C H. Multi-view subspace clustering network based on deep autoencoder[J]. Computer Engineering and Applications, 2020, 56(17): 60-68.
[14] LIU J, LIU X, YANG Y, et al. Multiview subspace clustering via co-training robust data representation[J]. IEEE Transa-ctions on Neural Networks and Learning Systems, 2022, 33(10): 5177-5189.
[15] LIU S Y, WANG S W, ZHANG P, et al. Efficient one-pass multi-view subspace clustering with consensus anchors[C]// Proceedings of the AAAI Conference on Artificial Intelligence, 2022: 7576-7584.
[16] LIU J L, WANG C, GAO J, et al. Multi-view clustering via joint nonnegative matrix factorization[C]//Proceedings of the 2013 SIAM International Conference on Data Mining. New York: SDM, 2013: 252-260.
[17] LI S Y, JIANG Y, ZHOU Z H. Partial multi-view clustering[C]//Proceedings of the AAAI Conference on Artificial Inte-lligence, 2014: 1968-1974.
[18] ZONG L, ZHANG X, ZHAO L, et al. Multi-view clustering via multi-manifold regularized non-negative matrix factori-zation[J]. Neural Networks, 2017, 88: 74-89.
[19] KHAN G A, HU J, LI T R, et al. Weighted multi-view data clustering via joint non-negative matrix factorization[C]//Proceedings of the 2019 IEEE 14th International Conference on Intelligent Systems and Knowledge Engineering. Piscataway: IEEE, 2019: 1159-1165.
[20] XIANG J S, LI C H, YANG H J, et al. Dual auto-weighted multi-view clustering via autoencoder-like nonnegative matrix factorization[J]. Information Sciences, 2024, 667: 1-18.
[21] 徐霜, 余琍. 利用正则化矩阵分解技术的多视图聚类方法[J]. 计算机工程与应用, 2019, 55(14): 142-147.
XU S, YU Li. Regularized matrix factorization algorithm for multi-view data clustering[J]. Computer Engineering and Applications, 2019, 55(14): 142-147.
[22] KANG Z, SHI G, HUANG S, et al. Multi-graph fusion for multi-view spectral clustering[J]. Knowledge-Based Systems, 2020, 189: 1-9.
[23] JIANG K B, ZHOU S B, QIAN X Z, et al. Dynamic-fusion multi-view projection clustering algorithm[J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(5): 1147-1156.
[24] 张衡, 谭晓阳, 金鑫. 基于多视图聚类的自然图像边缘检测[J]. 模式识别与人工智能, 2016, 29(2): 163-170.
ZHANG H, TAN X Y, JIN X. Multi-view clustering based natural image contour detection[J]. Pattern Recognition and Artificial Intelligence, 2016, 29(2): 163-170.
[25] LIU X W, DOU Y, YIN J P, et al. Multiple Kernel k-means clustering with matrix-induced regularization[C]//Proceedings of the 30th AAAI Conference on Artificial Intelligence, 2016: 1888-1894.
[26] XU J L, HAN J W, NIE F P. Discriminatively embedded k-means for multi-view clustering[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2016: 5356-5364.
[27] XU J L, HAN J W, NIE F P, et al. Re-weighted discriminatively embedded k-means for multiview clustering[J]. IEEE Transactions on Image Processing, 2017, 26(6): 3016-3027.
[28] WANG X B, LEI Z, GUO X J, et al. Multi-view subspace clustering with intactness-aware similarity[J]. Pattern Recognition, 2019, 88: 50-63.
[29] LEE D D, SEUNG H S. Algorithms for non?negative matrix factorization[C]//Proceedings of the 13th International Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2000: 535-541.
[30] CHEN M S, HUANG L, WANG C D, et al. Relaxed multiview clustering in latent embedding space[J]. Information Fusion, 2021, 68: 8-21.
[31] WAN X H, LIU X W, LIU J Y, et al. Auto-weighted multi-view clustering for large-scale data[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2023: 10078-10086.
[32] LEE D D, SEUNG H S. Learning the parts of objects by non-negative matrix factorization[J]. Nature, 1999, 401: 788-791.
[33] DING C, HE X F, SIMON H D. On the equivalence of nonnegative matrix factorization and spectral clustering[C]//Proceedings of the 2005 SIAM International Conference on Data Mining, 2005: 606-610.
[34] MASSON M H, DENOEUX T. ECM: an evidential version of the fuzzy C-means algorithm[J]. Pattern Recognition, 2007, 41(4): 1384-1397.
[35] NIE F P, LI J, LI X L, et al. Parameter-free auto-weighted multiple graph learning: a framework for multiview clustering and semi-supervised classific-ation[C]//Proceedings of International Joint Conference on Artificial Intelligence. New York: AAAI Press, 2016: 1881-1887.
[36] ZHAN K, ZHANG C, GUAN J, et al. Graph Learning for Multiview Clustering[J]. IEEE Transactions on Cybernetics, 2018, 48(10): 2887-2895.
[37] ZHAN K, NIE F P, WANG J, et al. Multiview consensus graph clustering[J]. IEEE Transactions on Image Processing,2018, 28(3): 1261-1270.
[38] TANG C, LI Z L, WANG J, et al. Unified one-step multi-view spectral clustering[J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 35(6): 6449-6460.
[39] ZHANG P, LIU X W, XIONG J, et al. Consensus one-step multi-view subspace clustering[J]. IEEE Transactions on Knowledge and Data Engineering, 2020, 34 (10): 4676-4689.
[40] WANG S W, LIU X W, ZHU X Z, et al. Fast parameter-free multi-view subspace clustering with consensus anchor guidance[J]. IEEE Transactions on Image Processing, 2021, 31: 556-568.