Spectral Clustering Based on k-Nearest Neighbor Graph

Abstract : Finding clusters in data is a challenging task when the clusters differ widely in shapes, sizes, and densities. We present a novel spectral algorithm Speclus with a similarity measure based on modified mutual nearest neighbor graph. The resulting affinity matrix reflex the true structure of data. Its eigenvectors, that do not change their sign, are used for clustering data. The algorithm requires only one parameter – a number of nearest neighbors, which can be quite easily established. Its performance on both artificial and real data sets is competitive to other solutions.
Type de document :
Communication dans un congrès
Agostino Cortesi; Nabendu Chaki; Khalid Saeed; Sławomir Wierzchoń. 11th International Conference on Computer Information Systems and Industrial Management (CISIM), Sep 2012, Venice, Italy. Springer, Lecture Notes in Computer Science, LNCS-7564, pp.254-265, 2012, Computer Information Systems and Industrial Management. 〈10.1007/978-3-642-33260-9_22〉
Liste complète des métadonnées

Littérature citée [19 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01551732
Contributeur : Hal Ifip <>
Soumis le : vendredi 30 juin 2017 - 14:43:22
Dernière modification le : lundi 22 janvier 2018 - 11:08:07
Document(s) archivé(s) le : lundi 22 janvier 2018 - 21:34:48

Fichier

978-3-642-33260-9_22_Chapter.p...
Fichiers produits par l'(les) auteur(s)

Licence


Distributed under a Creative Commons Paternité 4.0 International License

Identifiants

Citation

Małgorzata Lucińska, Sławomir Wierzchoń. Spectral Clustering Based on k-Nearest Neighbor Graph. Agostino Cortesi; Nabendu Chaki; Khalid Saeed; Sławomir Wierzchoń. 11th International Conference on Computer Information Systems and Industrial Management (CISIM), Sep 2012, Venice, Italy. Springer, Lecture Notes in Computer Science, LNCS-7564, pp.254-265, 2012, Computer Information Systems and Industrial Management. 〈10.1007/978-3-642-33260-9_22〉. 〈hal-01551732〉

Partager

Métriques

Consultations de la notice

129

Téléchargements de fichiers

384