Multiple graphs clustering by gradient flow method

Hong Zhu*, Chuan Chen, Lizhi Liao, Michael Kwok Po Ng

*Corresponding author for this work

Research output: Contribution to journalJournal articlepeer-review

9 Citations (Scopus)


The core issue of multiple graphs clustering is to find clusters of vertices from graphs such that these clusters are well-separated in each graph and clusters are consistent across different graphs. The problem can be formulated as a multiple orthogonality constrained optimization model which can be shown to be a relaxation of a multiple graphs cut problem. The resulting optimization problem can be solved by a gradient flow iterative method. The convergence of the proposed iterative scheme can be established. Numerical examples are presented to demonstrate the effectiveness of the proposed method for solving multiple graphs clustering problems in terms of clustering accuracy and computational efficiency.

Original languageEnglish
Pages (from-to)1819-1845
Number of pages27
JournalJournal of the Franklin Institute
Issue number4
Publication statusPublished - Mar 2018

Scopus Subject Areas

  • Control and Systems Engineering
  • Signal Processing
  • Computer Networks and Communications
  • Applied Mathematics


Dive into the research topics of 'Multiple graphs clustering by gradient flow method'. Together they form a unique fingerprint.

Cite this