Show simple item record

dc.contributor.authorTitterington, Mike
dc.contributor.authorRobert, Christian P.
dc.contributor.authorMarin, Jean-Michel
dc.contributor.authorCucala, Lionel
dc.date.accessioned2009-07-07T09:37:48Z
dc.date.available2009-07-07T09:37:48Z
dc.date.issued2009
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/867
dc.language.isoenen
dc.subjectBayesian inferenceen
dc.subjectBoltzmann modelen
dc.subjectCompatible conditionalsen
dc.subjectClassificationen
dc.subjectMCMC algorithmsen
dc.subjectPseudo-likelihooden
dc.subjectPath samplingen
dc.subjectPerfect samplingen
dc.subjectNormalising constanten
dc.subject.ddc519en
dc.titleA Bayesian reassessment of nearest-neighbour classificationen
dc.typeArticle accepté pour publication ou publié
dc.contributor.editoruniversityotherUniversity of Glasgow;Royaume-Uni
dc.contributor.editoruniversityotherINRIA – Université Paris Sud - Paris XI;France
dc.contributor.editoruniversityotherINSEE;France
dc.description.abstractenThe k-nearest-neighbour procedure is a well-known deterministic method used in supervised classification. This paper proposes a reassessment of this approach as a statistical technique derived from a proper probabilistic model; in particular, we modify the assessment made in a previous analysis of this method undertaken by Holmes and Adams (2002, 2003), and evaluated by Manocha and Girolami (2007), where the underlying probabilistic model is not completely well-defined. Once a clear probabilistic basis for the k-nearest-neighbour procedure is established, we derive computational tools for conducting Bayesian inference on the parameters of the corresponding model. In particular, we assess the difficulties inherent to pseudo-likelihood and to path sampling approximations of an intractable normalising constant, and propose a perfect sampling strategy to implement a correct MCMC sampler associated with our model. If perfect sampling is not available, we suggest using a Gibbs sampling approximation. Illustrations of the performance of the corresponding Bayesian classifier are provided for several benchmark datasets, demonstrating in particular the limitations of the pseudo-likelihood approximation in this set-up.en
dc.relation.isversionofjnlnameJournal of the American Statistical Association
dc.relation.isversionofjnlvol104
dc.relation.isversionofjnlissue485
dc.relation.isversionofjnldate2009-03
dc.relation.isversionofjnlpages263-273
dc.relation.isversionofdoihttp://dx.doi.org/10.1198/jasa.2009.0125
dc.identifier.urlsitehttp://hal.inria.fr/inria-00143783/en/en
dc.description.sponsorshipprivateouien
dc.relation.isversionofjnlpublisherAmerican Statistical Association
dc.subject.ddclabelProbabilités et mathématiques appliquéesen


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record