4.7 Article

Graph matching and clustering using spectral partitions

Journal

PATTERN RECOGNITION
Volume 39, Issue 1, Pages 22-34

Publisher

ELSEVIER SCI LTD
DOI: 10.1016/j.patcog.2005.06.014

Keywords

inexact graph matching; graph simplification; graph partition; graph spectral clustering; Fiedler vector

Ask authors/readers for more resources

Although inexact graph-matching is a problem of potentially exponential complexity, the problem may be simplified by decomposing the graphs to be matched into smaller subgraphs. If this is done, then the process may cast into a hierarchical framework and hence rendered suitable for parallel computation. In this paper we describe a spectral method which can be used to partition graphs into nonoverlapping subgraphs. In particular, we demonstrate how the Fiedler-vector of the Laplacian matrix can be used to decompose graphs into non-overlapping neighbourhoods that can be used for the purposes of both matching and clustering. (c) 2005 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.7
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available