A proposal for supervised clustering with Dirichlet Process using labels

Simple item page
datacite.alternateIdentifier.citationPATTERN RECOGNITION LETTERS,Vol.80,52-57,2016
datacite.alternateIdentifier.doi10.1016/j.patrec.2016.05.019
datacite.creatorPeralta Márquez, Billy
datacite.creatorCaro, Alberto
datacite.creatorSoto, Alvaro
datacite.date2016
datacite.subject.englishDirichlet Process
datacite.subject.englishSupervised clustering
datacite.subject.englishClustering
datacite.titleA proposal for supervised clustering with Dirichlet Process using labels
dc.date.accessioned2021-04-30T16:34:18Z
dc.date.available2021-04-30T16:34:18Z
dc.description.abstractSupervised clustering is an emerging area of machine learning, where the goal is to find class-uniform clusters. However, typical state-of-the-art algorithms use a fixed number of clusters. In this work, we propose a variation of a non-parametric Bayesian modeling for supervised clustering. Our approach consists of modeling the clusters as a mixture of Gaussians with the constraint of encouraging clusters of points with the same label. In order to estimate the number of clusters, we assume a-priori a countably infinite number of clusters using a variation of Dirichlet Process model over the prior distribution. In our experiments, we show that our technique typically outperforms the results of other clustering techniques. (C) 2016 Elsevier B.V. All rights reserved.
dc.identifier.urihttp://repositoriodigital.uct.cl/handle/10925/3027
dc.language.isoen
dc.publisherELSEVIER
dc.sourcePATTERN RECOGNITION LETTERS
oaire.resourceTypeArticle
uct.catalogadorWOS
uct.indizacionSCI
Files
Collections
Migración Web of Science