Translator Disclaimer
August 2008 Multiway Spectral Clustering: A Margin-Based Perspective
Zhihua Zhang, Michael I. Jordan
Statist. Sci. 23(3): 383-403 (August 2008). DOI: 10.1214/08-STS266

Abstract

Spectral clustering is a broad class of clustering procedures in which an intractable combinatorial optimization formulation of clustering is “relaxed” into a tractable eigenvector problem, and in which the relaxed solution is subsequently “rounded” into an approximate discrete solution to the original problem. In this paper we present a novel margin-based perspective on multiway spectral clustering. We show that the margin-based perspective illuminates both the relaxation and rounding aspects of spectral clustering, providing a unified analysis of existing algorithms and guiding the design of new algorithms. We also present connections between spectral clustering and several other topics in statistics, specifically minimum-variance clustering, Procrustes analysis and Gaussian intrinsic autoregression.

Citation

Download Citation

Zhihua Zhang. Michael I. Jordan. "Multiway Spectral Clustering: A Margin-Based Perspective." Statist. Sci. 23 (3) 383 - 403, August 2008. https://doi.org/10.1214/08-STS266

Information

Published: August 2008
First available in Project Euclid: 28 January 2009

zbMATH: 1329.62294
MathSciNet: MR2483910
Digital Object Identifier: 10.1214/08-STS266

Rights: Copyright © 2008 Institute of Mathematical Statistics

JOURNAL ARTICLE
21 PAGES


SHARE
Vol.23 • No. 3 • August 2008
Back to Top