Hyper-Laplacian Regularized Concept Factorization in Low-Rank Tensor Space for Multi-View Clustering

被引:0
作者
Yu, Zixiao [1 ]
Fu, Lele [2 ]
Chen, Yongyong [3 ,4 ]
Cai, Zhiling [5 ]
Chao, Guoqing [6 ]
机构
[1] Univ Melbourne, Fac Engn & Informat Technol, Melbourne 3052, Australia
[2] Sun Yat Sen Univ, Sch Syst Sci & Engn, Guangzhou 510006, Peoples R China
[3] Harbin Inst Technol Shenzhen, Shenzhen Key Lab Visual Object Detect & Recognit, Shenzhen 518055, Peoples R China
[4] Guangdong Prov Key Lab Novel Secur Intelligence T, Shenzhen 518055, Peoples R China
[5] Fujian Agr & Forestry Univ, Coll Comp & Informat Sci, Fuzhou 350002, Peoples R China
[6] Harbin Inst Technol Weihai, Sch Comp Sci & Technol, Weihai 264209, Peoples R China
来源
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE | 2024年
基金
中国国家自然科学基金;
关键词
Tensors; Matrix decomposition; Electronic mail; Correlation; Feature extraction; Computational intelligence; Optimization; Multi-view clustering; concept factorization; self-weighted tensor Schatten p-norm;
D O I
10.1109/TETCI.2024.3449920
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Tensor-oriented multi-view subspace clustering has achieved significant strides in assessing high-order correlations of multi-view data. Nevertheless, most of existing investigations are typically hampered by the two flaws: (1) Self-representation based tensor subspace learning usually induces high time and space complexity, and is limited in perceiving nonlinear local structure in the embedding space. (2) The tensor singular value decomposition model redistributes each singular value equally without considering the diverse importance among them. To well cope with the above issues, we propose a hyper-Laplacian regularized concept factorization (HLRCF) in low-rank tensor space for multi-view clustering. Specifically, HLRCF adopts the concept factorization to explore the latent cluster-wise representation of each view. Further, the hypergraph Laplacian regularization endows the model with the capability of extracting the nonlinear local structures in the latent space. Considering that different tensor singular values associate structural information with unequal importance, we develop a self-weighted tensor Schatten p-norm to constrain the tensor comprised of all cluster-wise representations. Notably, the tensor with smaller size greatly decreases the time and space complexity in the low-rank optimization. Finally, experimental results on eight benchmark datasets exhibit that HLRCF outperforms other multi-view methods, showing its superior performance.
引用
收藏
页数:15
相关论文
共 50 条
  • [1] Multi-view clustering via deep concept factorization
    Chang, Shuai
    Hu, Jie
    Li, Tianrui
    Wang, Hao
    Peng, Bo
    [J]. KNOWLEDGE-BASED SYSTEMS, 2021, 217
  • [2] Concept Factorization Based Multiview Clustering for Large-Scale Data
    Chen, Man-Sheng
    Wang, Chang-Dong
    Huang, Dong
    Lai, Jian-Huang
    Yu, Philip S.
    [J]. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2024, 36 (11) : 5784 - 5796
  • [3] Chen MS, 2020, AAAI CONF ARTIF INTE, V34, P3513
  • [4] Neighbor-aware deep multi-view clustering via graph convolutional network
    Du, Guowang
    Zhou, Lihua
    Li, Zhongxue
    Wang, Lizhen
    Lu, Kevin
    [J]. INFORMATION FUSION, 2023, 93 : 330 - 343
  • [5] NUCLEAR NORM OF HIGHER-ORDER TENSORS
    Friedland, Shmuel
    Lim, Lek-Heng
    [J]. MATHEMATICS OF COMPUTATION, 2018, 87 (311) : 1255 - 1281
  • [6] Unified Low-Rank Tensor Learning and Spectral Embedding for Multi-View Subspace Clustering
    Fu, Lele
    Chen, Zhaoliang
    Chen, Yongyong
    Wang, Shiping
    [J]. IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 25 : 4972 - 4985
  • [7] Low-rank tensor approximation with local structure for multi-view intrinsic subspace clustering
    Fu, Lele
    Yang, Jinghua
    Chen, Chuan
    Zhang, Chuanfu
    [J]. INFORMATION SCIENCES, 2022, 606 : 877 - 891
  • [8] An overview of recent multi-view clustering
    Fu, Lele
    Lin, Pengfei
    Vasilakos, Athanasios V.
    Wang, Shiping
    [J]. NEUROCOMPUTING, 2020, 402 (402) : 148 - 161
  • [9] Gao QX, 2020, AAAI CONF ARTIF INTE, V34, P3930
  • [10] Enhanced Tensor RPCA and its Application
    Gao, Quanxue
    Zhang, Pu
    Xia, Wei
    Xie, Deyan
    Gao, Xinbo
    Tao, Dacheng
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2021, 43 (06) : 2133 - 2140