Evidential clustering of large dissimilarity data - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Knowledge-Based Systems Année : 2016

Evidential clustering of large dissimilarity data

Résumé

In evidential clustering, the membership of objects to clusters is considered to be uncertain and is represented by Dempster-Shafer mass functions, forming a credal partition. The EVCLUS algorithm constructs a credal partition in such a way that larger dissimilarities between objects correspond to higher degrees of conflict between the associated mass functions. In this paper, we present several improvements to EVCLUS, making it applicable to very large dissimilarity data. First, the gradient-based optimization procedure in the original EVCLUS algorithm is replaced by a much faster iterative row-wise quadratic programming method. Secondly, we show that EVCLUS can be provided with only a random sample of the dissimilarities, reducing the time and space complexity from quadratic to roughly linear. Finally, we introduce a two-step approach to construct credal partitions assigning masses to selected pairs of clusters, making the algorithm outputs more informative than those of the original EVCLUS, while remaining manageable for large numbers of clusters.
Fichier principal
Vignette du fichier
evclus_kbs_v2_clean.pdf (1.35 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01324491 , version 1 (01-06-2016)

Identifiants

Citer

Thierry Denoeux, Songsak Sriboonchitta, Orakanya Kanjanatarakul. Evidential clustering of large dissimilarity data. Knowledge-Based Systems, 2016, 106, pp.179-195. ⟨10.1016/j.knosys.2016.05.043⟩. ⟨hal-01324491⟩
70 Consultations
219 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More