Spectral Clustering Based on Analysis of Eigenvector Properties - Computer Information Systems and Industrial Management Access content directly
Conference Papers Year : 2014

Spectral Clustering Based on Analysis of Eigenvector Properties

Abstract

In this paper we propose a new method for choosing the number of clusters and the most appropriate eigenvectors, that allow to obtain the optimal clustering. To accomplish the task we suggest to examine carefully properties of adjacency matrix eigenvectors: their weak localization as well as the sign of their values. The algorithm has only one parameter — the number of mutual neighbors. We compare our method to several clustering solutions using different types of datasets. The experiments demonstrate that our method outperforms in most cases many other clustering algorithms.
Fichier principal
Vignette du fichier
978-3-662-45237-0_6_Chapter.pdf (177.23 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01405553 , version 1 (30-11-2016)

Licence

Attribution

Identifiers

Cite

Małgorzata Lucińska, Sławomir T. Wierzchoń. Spectral Clustering Based on Analysis of Eigenvector Properties. 13th IFIP International Conference on Computer Information Systems and Industrial Management (CISIM), Nov 2014, Ho Chi Minh City, Vietnam. pp.43-54, ⟨10.1007/978-3-662-45237-0_6⟩. ⟨hal-01405553⟩
63 View
1464 Download

Altmetric

Share

Gmail Facebook X LinkedIn More