Breaking down multi-view clustering: A comprehensive review of multi-view approaches for complex data structures

被引:9
作者
Haris, Muhammad [1 ,2 ]
Yusoff, Yusliza [1 ]
Zain, Azlan Mohd [1 ]
Khattak, Abid Saeed [1 ,2 ]
Hussain, Syed Fawad [3 ]
机构
[1] Univ Teknol Malaysia, Fac Comp, Skudai 81310, Johor, Malaysia
[2] Khushal Khan Khattak Univ, Dept Comp Sci & Bioinformat, Karak, Pakistan
[3] Univ Birmingham, Sch Comp Sci, Birmingham, England
关键词
Multi -view subspace clustering; Non -negative matrix factorization; Multi -view matrix factorization; Subspace learning; Model -based clustering; Multi -view unsupervised deep learning; NONNEGATIVE MATRIX FACTORIZATION; SUBSPACE ANALYSIS; SPARSE; ALGORITHM; FRAMEWORK;
D O I
10.1016/j.engappai.2024.107857
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Multi-View Clustering (MVC) is an emerging research area aiming to cluster multiple views of the same data, which has recently drawn substantial attention. Various approaches have been proposed in the literature, from a unified objective function across views, transfer learning approaches, and early, intermediate, and late fusion strategies to combine data emanating from multiple views. Despite the increasing popularity of MVC methods, there is still a lack of systematic categorization for MVC methodologies. Existing literature reviews on MVC often overlook the importance of providing a cohesive classification, and deep learning has significantly impacted the field of MVC by effectively analyzing complex data structures. This research addresses the need for consolidation by classifying MVC approaches into two main types: generative and discriminative. The Discriminative approaches are further split into four subclasses, including a category specifically focused on deep learning-based methods. This study emphasizes the growing importance of deep learning in MVC approaches. By categorizing deep learning as a separate class of algorithms, we intend to emphasize its significant influence on the MVC. A systematic comparison among different classes of MVC algorithms on benchmark textual and image datasets was conducted to objectively assess the efficacy of different MVC methodologies. Therefore, this review comprehensively analyzed existing MVC paradigms and identified potential research directions.
引用
收藏
页数:21
相关论文
共 118 条
[11]   Multi-view clustering via deep concept factorization [J].
Chang, Shuai ;
Hu, Jie ;
Li, Tianrui ;
Wang, Hao ;
Peng, Bo .
KNOWLEDGE-BASED SYSTEMS, 2021, 217
[12]  
Chao GQ, 2018, Arxiv, DOI arXiv:1712.06246
[13]  
Chao Guoqing, 2021, IEEE Trans Artif Intell, V2, P146, DOI 10.1109/tai.2021.3065894
[14]   Discriminative K-Means Laplacian Clustering [J].
Chao, Guoqing .
NEURAL PROCESSING LETTERS, 2019, 49 (01) :393-405
[15]   Joint Shared-and-Specific Information for Deep Multi-View Clustering [J].
Chen, Jin ;
Huang, Aiping ;
Gao, Wei ;
Niu, Yuzhen ;
Zhao, Tiesong .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2023, 33 (12) :7224-7235
[16]  
Chen MS, 2020, AAAI CONF ARTIF INTE, V34, P3513
[17]   Diversity embedding deep matrix factorization for multi-view clustering [J].
Chen, Zexi ;
Lin, Pengfei ;
Chen, Zhaoliang ;
Ye, Dongyi ;
Wang, Shiping .
INFORMATION SCIENCES, 2022, 610 :114-125
[18]   Hierarchical Clustering: Objective Functions and Algorithms [J].
Cohen-Addad, Vincent ;
Kanade, Varun ;
Mallmann-Trenn, Frederik ;
Mathieu, Claire .
JOURNAL OF THE ACM, 2019, 66 (04)
[19]   MAXIMUM LIKELIHOOD FROM INCOMPLETE DATA VIA EM ALGORITHM [J].
DEMPSTER, AP ;
LAIRD, NM ;
RUBIN, DB .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL, 1977, 39 (01) :1-38
[20]   Multi-view clustering guided by unconstrained non-negative matrix factorization [J].
Deng, Ping ;
Li, Tianrui ;
Wang, Dexian ;
Wang, Hongjun ;
Peng, Hong ;
Horng, Shi-Jinn .
KNOWLEDGE-BASED SYSTEMS, 2023, 266