A. Ng, M. Jordan, and Y. Weiss. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS, page 849--856. MIT Press, (2001)
Abstract
Despite many empirical successes of spectral
clustering methods -- algorithms that cluster points
using eigenvectors of matrices derived from the
distances between the points -- there are several
unresolved issues. First, there is 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.
%0 Conference Paper
%1 ng-spectral-clustering-analysis-2002
%A Ng, Andrew Y.
%A Jordan, Michael I.
%A Weiss, Yair
%B ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS
%D 2001
%I MIT Press
%K clustering spectral
%P 849--856
%T On Spectral Clustering: Analysis and an algorithm
%U http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.19.8100
%X Despite many empirical successes of spectral
clustering methods -- algorithms that cluster points
using eigenvectors of matrices derived from the
distances between the points -- there are several
unresolved issues. First, there is 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.
@inproceedings{ng-spectral-clustering-analysis-2002,
abstract = {Despite many empirical successes of spectral
clustering methods -- algorithms that cluster points
using eigenvectors of matrices derived from the
distances between the points -- there are several
unresolved issues. First, there is 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.},
added-at = {2016-07-12T19:24:18.000+0200},
author = {Ng, Andrew Y. and Jordan, Michael I. and Weiss, Yair},
biburl = {https://www.bibsonomy.org/bibtex/2e9c06dab81a9a2e06123cd9b31d3d83f/mhwombat},
booktitle = {ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS},
description = {On Spectral Clustering: Analysis and an algorithm},
interhash = {b72c97e659127fc653a0d51143d85b0c},
intrahash = {e9c06dab81a9a2e06123cd9b31d3d83f},
keywords = {clustering spectral},
pages = {849--856},
publisher = {MIT Press},
timestamp = {2016-07-12T19:25:30.000+0200},
title = {On Spectral Clustering: Analysis and an algorithm},
url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.19.8100},
year = 2001
}