On Spectral Clustering: Analysis and an algorithm

Abstract

Yair Weiss School of CS & Engr. The Hebrew Univ. yweiss@cs.huji.ac.il Despite many empirical successes of spectral clustering methodsalgorithms that cluster points using eigenvectors of matrices derived from the datathere are several unresolved issues. First , there are a wide variety of algorithms that use the eigenvectors in slightly different ways. Second, many of these algorithms have no proof that they will actually compute a reasonable clustering. In this paper, we present a simple spectral clustering algorithm that can be implemented using a few lines of Matlab. Using tools from matrix perturbation theory, we analyze the algorithm, and give conditions under which it can be expected to do well. We also show surprisingly good experimental results on a number of challenging clustering problems.

Extracted Key Phrases

1 Figure or Table

0200400600'03'05'07'09'11'13'15'17
Citations per Year

6,326 Citations

Semantic Scholar estimates that this publication has 6,326 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{Ng2001OnSC, title={On Spectral Clustering: Analysis and an algorithm}, author={Andrew Y. Ng and Michael I. Jordan and Yair Weiss}, booktitle={NIPS}, year={2001} }