Consistency  of spectral clustering

Consistency of spectral clustering (2004)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Ulrike von Luxburg , Mikhail Belkin , Olivier Bousquet

Citations:

170 - 11 self

BibTeX

@INPROCEEDINGS{Luxburg04consistencyof,
    author = {Ulrike von Luxburg and Mikhail Belkin and Olivier Bousquet},
    title = {Consistency of spectral clustering},
    booktitle = {},
    year = {2004},
    pages = {857--864},
    publisher = {MIT Press}
}

Years of Citing Articles

Bookmark

OpenURL

Abstract

Consistency is a key property of statistical algorithms, when the data is drawn from some underlying probability distribution. Surprisingly, despite decades of work, little is known about consistency of most clustering algorithms. In this paper we investigate consistency of a popular family of spectral clustering algorithms, which cluster the data with the help of eigenvectors of graph Laplacian matrices. We show that one of the two of major classes of spectral clustering (normalized clustering) converges under some very general conditions, while the other (unnormalized), is only consistent under strong additional assumptions, which, as we demonstrate, are not always satisfied in real data. We conclude that our analysis provides strong evidence for the superiority of normalized spectral clustering in practical applications. We believe that methods used in our analysis will provide a basis for future exploration of Laplacian-based methods in a statistical setting.

Citations

1824	Normalized cuts and image segmentation - Shi, Malik - 2000
1523	Matrix Computations - Golub, Loan - 1983
1430	The Elements of Statistical Learning - Hastie, Tibshirani, et al. - 2009
1305	Perturbation Theory for Linear Operators - Kato - 1980
912	Data clustering: a review - Jain, Murty, et al. - 1999
757	On spectral clustering: Analysis and an algorithm - Ng, Jordan, et al.
710	Spectral Graph Theory - Chung - 1997
519	Laplacian eigenmaps for dimensionality reduction and data representation - Belkin, Niyogi - 2003
471	Matrix Perturbation Theory - Stewart, Sun - 1990
457	Convergence of Stochastic Processes - Pollard - 1984
437	Partitioning sparse matrices with eigenvectors of graphs - Pothen, Simon, et al. - 1990
315	Matrix Analysis - Bhatia - 1997
294	Algebraic connectivity of graphs - Fiedler - 1973
286	Learning with Local and Global Consistency - Zhou, Bousquet, et al. - 2004
262	Partitioning of unstructured problems for parallel processing - Simon - 1991
261	Segmentation using eigenvectors: A unifying view - Weiss
239	Weak Convergence and Empirical Processes - Vaart, Wellner - 1996
239	Markov Chains - Norris - 1996
220	Co-clustering documents and words using bipartite spectral graph partitioning - Dhillon - 2001
216	Markov Chains: Gibbs Fields, Monte Carlo Simulation, and Queues - Bremaud - 1999
203	Vetta A: On clustering: good, bad and spectral - Kannan, Vampala
175	Expander flows, geometric embeddings and graph partitioning - Arora, Rao, et al. - 2004
173	Random walks on graphs: A survey - Lovasz - 1996
172	Model-based clustering, discriminant analysis, and density estimation - Fraley, Raftery - 2002
171	Semi-Supervised Learning - Chapelle, Schlkopf, et al. - 2006
170	New spectral methods for ratio cut partitioning and clustering - Hagen, Kahng - 1992
168	T: Estimating the Number of Clusters in a Dataset via the Gap Statistic - Tibshirani, Walther, et al. - 2000
160	An Improved Spectral Graph Partitioning Algorithm for Mapping Parallel Computations - Hendrickson, Leland
159	lower bound for the smallest eigenvalue of the Laplacian - Cheeger, “A - 1970
153	Transductive learning via spectral graph partitioning - Joachims - 2003
130	J: A Random Walks View of Spectral Segmentation - Meila, Shi
129	Functional Analysis - Reed, Simon - 1980
127	Lower bounds for partitioning of graphs - Donath, Homan - 1973
124	Spectral partitioning works: Planar graphs and finite element meshes - Spielman, Teng - 1996
120	Cluster Kernels for Semi-Supervised Learning - Chapelle, Weston, et al.
116	Semi-supervised learning on Riemannian manifolds - Belkin, Niyogi
115	A min-max cut algorithm for graph partitioning and data clustering - Ding, He, et al. - 2001
113	The Laplacian spectrum of graphs - Mohar
102	Handbook of Matrices - Lutkepohl - 1996
93	F: A simple min-cut algorithm - Stoer, Wagner - 1997
93	Methodologies in spectral analysis of large dimensional random matrices - Bai - 1999
85	Spectral graph theory, volume 92 - Chung - 1997
79	Spectral Approximations of Linear Operators - CHATELIN - 1983
73	A stability based method for discovering structure in clustered data - Ben-Hur, Elisseeff, et al. - 2002
70	Some applications of Laplace eigenvalues of graphs - Mohar
66	The Numerical Treatment of Integral Equations - Baker - 1977
65	Learning Spectral Clustering - Bach, Jordan - 2004
63	Towards a theoretical foundation for laplacian-based manifold methods - Belkin, Niyogi - 2005
58	Finding good approximate vertex and edge partitions is NP-hard - Bui, Jones - 1992
57	Spectral partitioning, the more eigenvectors the better - Alpert, Yao - 1995