Adaptive and structured graph learning for semi-supervised clustering

作者：

Highlights：

• construct a pairwise constraints indicator to ensure that all samples meet the pairwise constraints softly.

• Learn a dynamic graph adaptively and make sure that the learned graph contains a set of c clusters.

• Link the learned adaptive graph to the pairwise constraints indicator through an constraint of alignment to guide each other to optimize.

摘要

•construct a pairwise constraints indicator to ensure that all samples meet the pairwise constraints softly.•Learn a dynamic graph adaptively and make sure that the learned graph contains a set of c clusters.•Link the learned adaptive graph to the pairwise constraints indicator through an constraint of alignment to guide each other to optimize.

论文关键词：Structured graph learning,Semi-supervised clustering,Pairwise constraints

论文评审过程：Received 9 December 2021, Revised 27 March 2022, Accepted 19 April 2022, Available online 6 May 2022, Version of Record 6 May 2022.

论文官网地址：https://doi.org/10.1016/j.ipm.2022.102949