Deutsch
 
Hilfe Datenschutzhinweis Impressum
  DetailsucheBrowse

Datensatz

DATENSATZ AKTIONENEXPORT

Freigegeben

Bericht

Consistency of Spectral Clustering

MPG-Autoren
/persons/resource/persons76237

von Luxburg,  U
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Max Planck Institute for Biological Cybernetics, Max Planck Society;

/persons/resource/persons83824

Bousquet,  O
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Max Planck Institute for Biological Cybernetics, Max Planck Society;

Externe Ressourcen
Es sind keine externen Ressourcen hinterlegt
Volltexte (beschränkter Zugriff)
Für Ihren IP-Bereich sind aktuell keine Volltexte freigegeben.
Volltexte (frei zugänglich)

MPIK-TR-134.pdf
(Verlagsversion), 327KB

Ergänzendes Material (frei zugänglich)
Es sind keine frei zugänglichen Ergänzenden Materialien verfügbar
Zitation

von Luxburg, U., Belkin, M., & Bousquet, O.(2004). Consistency of Spectral Clustering (134). Tübingen, Germany: Max Planck Institute for Biological Cybernetics.


Zitierlink: https://hdl.handle.net/11858/00-001M-0000-0013-D74B-6
Zusammenfassung
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.