Contrastive Consensus Graph Learning for Multi-View Clustering

Dear Editor, This letter proposes a contrastive consensus graph learning model for multi-view clustering. Graphs are usually built to outline the correlation between multi-model objects in clustering task, and multiview aims learn that integrates spatial property of each view. Nevertheless, most graph-based models merely consider overall structure from all views but neglect local consistency diverse views, resulting lack global learned graph. To overcome this issue, deep convolutional network is explore latent information raw affinity graphs. Specifically, we employ constraint preserve Furthermore, reconstruction loss introduced achieve sample-level approximation reconstructed graphs graphs, which facilitates enhance learning. Experiments on six classical datasets demonstrate proposed outperforms other nine state-of-the-art algorithms.