Incomplete multi-view clustering via structure exploration and missing-view inference

Ziyu Wang, Lusi Li, Xin Ning, Wenkai Tan, Yongxin Liu, Houbing Song

Abstract: Highlights•Propose a novel unified approach named SEMI that can handle incomplete multi-view clustering effectively.•Jointly explore global, local, consistent, and inconsistent structures.•Infer missing views under the guidance of the explored structures.•The mutual reinforcement between multi-structure exploration and missing-view inference can be achieved.