Incomplete Multi-view Clustering via Structured Graph Learning

In real applications, multi-view clustering with incomplete data has played an important role in the data mining field. How to design an algorithm to promote the clustering performance is a challenging problem. In this paper, we propose an approach with learned graph to handle the case that each view suffers from some missing information. It combines incomplete multi-view data and clusters it simultaneously by learning the ideal structures. For each view, with an initial input graph, it excavates a clustering structure with the consideration of consistency with the other views. The learned structured graphs have exactly c (the predefined number of clusters) connected components so that the clustering results can be obtained without requiring any post-clustering. An efficient optimization strategy is provided, which can simultaneously handle both the whole and the partial regularization problems. The proposed method exhibits impressive performance in experiments.