Abstract
A partitioning of a set of n items is a grouping of these items into k disjoint, equally sized classes. Any partition can be modeled as a graph. The items become the vertices of the graph and two vertices are connected by an edge if and only if the associated items belong to the same class. In a planted partition model a graph that models a partition is given, which is obscured by random noise, i.e., edges within a class can get removed and edges between classes can get inserted. The task is to reconstruct the planted partition from this graph. We design a spectral partitioning algorithm and analyze how many items it misclassifies in the worst case. The number of classes k is one parameter in the model that allows to control the difficulty of the problem. Our analysis extends the range of k for which any non-trivial quality guarantees can be given.
Partly supported by the Swiss National Science Foundation under the grant “Non-linear manifold learning”.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Alon, N., Krivelevich, M., Sudakov, B.: Finding a large hidden clique in a random graph. Random Structures and Algorithms 13, 457–466 (1998)
Bollobas, B., Scott, A.D.: Max cut for random graphs with a planted partition. Combinatorics, Probability and Computing 13, 451–474 (2004)
Boppana, R.B.: Eigenvalues and graph bisection: An average-case analysis. In: Proceedings of 28th IEEE Symposium on Foundations on Computer Science, pp. 280–285 (1987)
Condon, A., Karp, R.: Algorithms for graph partitioning on the planted partition model. Random Structures and Algorithms 8(2), 116–140 (1999)
Füredi, Z., Komlós, J.: The eigenvalues of random symmetric matrices. Combinatorica I 3, 233–241 (1981)
Garey, M.R., Johnson, D.S., Stockmeyer, L.: Some simplified np-complete graph problems. Theoretical Computer Science 1, 237–267 (1976)
Krivelevich, M., Vu, V.H.: On the concentration of eigenvalues of random symmetric matrices. Microsoft Technical Report, 60 (2000)
McSherry, F.: Spectral partitioning of random graphs. In: Proceedings of 42nd IEEE Symosium on Foundations of Computer Science, pp. 529–537 (2001)
Meila, M., Verma, D.: A comparison of spectral clustering algorithms. UW CSE Technical report 03-05-01
Shamir, R., Tsur, D.: Improved algorithms for the random cluster graph model. In: Proceedings 7th Scandinavian Workshop on Algorithm Theory, pp. 230–259 (2002)
Spielman, D., Teng, S.-H.: Spectral partitioning works: Planar graphs and finite element meshes. In: Proceedings of 37th IEEE Symposium on Foundations on Computer Science, pp. 96–105 (1996)
Stewart, G., Sun, J.: Matrix perturbation theory. Academic Press, Boston (1990)
Vu, V.H.: Spectral norm of random matrices. In: STOC 2005: Proceedings of the thirty-seventh annual ACM symposium on Theory of computing, pp. 423–430. ACM Press, New York (2005)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Giesen, J., Mitsche, D. (2005). Bounding the Misclassification Error in Spectral Partitioning in the Planted Partition Model. In: Kratsch, D. (eds) Graph-Theoretic Concepts in Computer Science. WG 2005. Lecture Notes in Computer Science, vol 3787. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11604686_36
Download citation
DOI: https://doi.org/10.1007/11604686_36
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-31000-6
Online ISBN: 978-3-540-31468-4
eBook Packages: Computer ScienceComputer Science (R0)