1092 articles – 619 Notices  [english version]
HAL : hal-00631620, version 1

Fiche détaillée  Récupérer au format
Pattern Recognition 44, 7 (2011) 1372-1386
Automatically finding clusters in normalized cuts
Mariano Tepper 1, Pablo Musé 2, A. Almansa 3, Marta Mejail 1

Normalized Cuts is a state-of-the-art spectral method for clustering. By applying spectral techniques, the data becomes easier to cluster and then k-means is classically used. Unfortunately the number of clusters must be manually set and it is very sensitive to initialization. Moreover, k-means tends to split large clusters, to merge small clusters, and to favor convex-shaped clusters. In this work we present a new clustering method which is parameterless, independent from the original data dimensionality and from the shape of the clusters. It only takes into account inter-point distances and it has no random steps. The combination of the proposed method with normalized cuts proved successful in our experiments.

1 :  Department of Computer Science [Buenos Aires]
University of Buenos Aires
2 :  Universidad de la Republica [Montevideo]
Universidad de la Republica Montevideo
3 :  Laboratoire Traitement et Communication de l'Information [Paris] (LTCI)
Télécom ParisTech – CNRS : UMR5141
Informatique/Traitement des images
Clustering – A contrario – Normalized cuts
Liste des fichiers attachés à ce document : 
clustering-PR2010.pdf(5.6 MB)