Self-weighted spectral clustering with parameter-free constraint

Rui Zhang, Feiping Nie, Xuelong Li

Research output: Contribution to journalArticlepeer-review

26 Scopus citations

Abstract

The constrained spectral clustering (or known as the semi-supervised spectral clustering) focuses on enhancing the clustering capability by utilizing the side information. In this paper, a novel constrained spectral clustering method is proposed based on deriving a sparse parameter-free similarity. Different from other works, the proposed method transforms the given pairwise constraints into the intrinsic graph similarity and the penalty graph similarity respectively instead of incorporating them into one single similarity. Besides, the optimal weight can be automatically achieved to balance the graph optimization problems between the intrinsic graph and the penalty graph. Equipped with a general framework of efficiently unraveling the bi-objective optimization, the proposed method could obtain both ratio cut and normalized cut clusterings via updating the weighted Laplacian matrix until convergence. Moreover, the proposed method is equivalent to the spectral clustering, when no side information is provided. Consequently, the effectiveness and the superiority of the proposed method are further verified both analytically and empirically.

Original languageEnglish
Pages (from-to)164-170
Number of pages7
JournalNeurocomputing
Volume241
DOIs
StatePublished - 7 Jun 2017

Keywords

  • Constrained spectral clustering
  • Parameter-free similarity
  • Quadratic weighted optimization

Fingerprint

Dive into the research topics of 'Self-weighted spectral clustering with parameter-free constraint'. Together they form a unique fingerprint.

Cite this