Power Spectral Clustering

Abstract : The problem of clustering has been an important problem since the early 20th century and several possible solutions were proposed. With the rise of computing machines clustering has become an important part of many data mining tasks, focussed on fast implementations. An important task related to clustering is image segmentation. In the set of solutions to the clustering problem, the method of spectral clustering has obtained wide interest due to its ability to detect non-convex clusters in the data. In this article, we propose a fast alternative to the spectral clustering, obtained by taking the Γ−limit. We explore the links between the new method and MST based clustering. We then show that the proposed method is as good as the spectral clustering with the help of experiments on several datasets. We also show that the new method is scalable to large data unlike the classical spectral clustering methods.
Type de document :
Pré-publication, Document de travail
2018
Liste complète des métadonnées

Littérature citée [37 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01516649
Contributeur : Aditya Challa <>
Soumis le : lundi 22 janvier 2018 - 12:18:20
Dernière modification le : samedi 28 juillet 2018 - 15:48:01
Document(s) archivé(s) le : jeudi 24 mai 2018 - 07:18:08

Fichier

Paper_PowerSpectral.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01516649, version 3

Citation

Aditya Challa, Sravan Danda, B S Daya Sagar, Laurent Najman. Power Spectral Clustering. 2018. 〈hal-01516649v3〉

Partager

Métriques

Consultations de la notice

215

Téléchargements de fichiers

213