Representation Learning in Multi-view Clustering: A Literature Review

Abstract Multi-view clustering (MVC) has attracted more and attention in the recent few years by making full use of complementary consensus information between multiple views to cluster objects into different partitions. Although there have been two existing works for MVC survey, neither them jointly takes popular deep learning-based methods consideration. Therefore, this paper, we conduct a comprehensive survey from perspective representation learning. It covers quantity multi-view including models, providing novel taxonomy algorithms. Furthermore, can be mainly divided categories, i.e., shallow MVC, where models are capable handling complex data structure as well showing better expression. In category, according means learning, further split it groups, graph subspace clustering. To comprehensive, basic research materials provided readers, containing introductions commonly used datasets with download link open source code library. end, some problems pointed out investigation development.