Tensorized Incomplete Multi-View Clustering with Intrinsic Graph Completion

Most of the existing incomplete multi-view clustering (IMVC) methods focus on attaining a consensus representation from different views but ignore important information hidden in missing and latent intrinsic structures each view. To tackle these issues, this paper, unified novel framework, named tensorized with graph completion (TIMVC_IGC) is proposed. Firstly, owing to effectiveness low-rank revealing inherent structure data, we exploit it infer instances construct complete for Afterwards, inspired by structural consistency, between-view consistency constraint imposed guarantee similarity graphs views. More importantly, TIMVC_IGC simultaneously learns explores correlations manifold sub-space using tensor constraint, such that can be obtained. Finally, sample gained co-regularization term final clustering. Experimental results several real-world databases illustrates proposed method outperform other state-of-the-art related