On Bayesian analysis of a finite generalized Dirichlet mixture via a Metropolis-within-Gibbs sampling
Identifieur interne : 004345 ( Main/Exploration ); précédent : 004344; suivant : 004346On Bayesian analysis of a finite generalized Dirichlet mixture via a Metropolis-within-Gibbs sampling
Auteurs : Nizar Bouguila [Canada] ; Djemel Ziou [Canada] ; Riad I. Hammoud [États-Unis]Source :
- Pattern Analysis and Applications [ 1433-7541 ] ; 2009-06-01.
English descriptors
- KwdEn :
Abstract
Abstract: In this paper, we present a fully Bayesian approach for generalized Dirichlet mixtures estimation and selection. The estimation of the parameters is based on the Monte Carlo simulation technique of Gibbs sampling mixed with a Metropolis-Hastings step. Also, we obtain a posterior distribution which is conjugate to a generalized Dirichlet likelihood. For the selection of the number of clusters, we used the integrated likelihood. The performance of our Bayesian algorithm is tested and compared with the maximum likelihood approach by the classification of several synthetic and real data sets. The generalized Dirichlet mixture is also applied to the problems of IR eye modeling and introduced as a probabilistic kernel for Support Vector Machines.
Url:
DOI: 10.1007/s10044-008-0111-4
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 003315
- to stream Istex, to step Curation: 003273
- to stream Istex, to step Checkpoint: 000D73
- to stream Main, to step Merge: 004456
- to stream Main, to step Curation: 004345
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">On Bayesian analysis of a finite generalized Dirichlet mixture via a Metropolis-within-Gibbs sampling</title><author><name sortKey="Bouguila, Nizar" sort="Bouguila, Nizar" uniqKey="Bouguila N" first="Nizar" last="Bouguila">Nizar Bouguila</name></author><author><name sortKey="Ziou, Djemel" sort="Ziou, Djemel" uniqKey="Ziou D" first="Djemel" last="Ziou">Djemel Ziou</name></author><author><name sortKey="Hammoud, Riad I" sort="Hammoud, Riad I" uniqKey="Hammoud R" first="Riad I." last="Hammoud">Riad I. Hammoud</name></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:D65DF631C50E3E2230F33877937834B44BE7F510</idno><date when="2008" year="2008">2008</date><idno type="doi">10.1007/s10044-008-0111-4</idno><idno type="url">https://api.istex.fr/ark:/67375/VQC-NTLG254N-D/fulltext.pdf</idno><idno type="wicri:Area/Istex/Corpus">003315</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">003315</idno><idno type="wicri:Area/Istex/Curation">003273</idno><idno type="wicri:Area/Istex/Checkpoint">000D73</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Checkpoint">000D73</idno><idno type="wicri:doubleKey">1433-7541:2008:Bouguila N:on:bayesian:analysis</idno><idno type="wicri:Area/Main/Merge">004456</idno><idno type="wicri:Area/Main/Curation">004345</idno><idno type="wicri:Area/Main/Exploration">004345</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">On Bayesian analysis of a finite generalized Dirichlet mixture via a Metropolis-within-Gibbs sampling</title><author><name sortKey="Bouguila, Nizar" sort="Bouguila, Nizar" uniqKey="Bouguila N" first="Nizar" last="Bouguila">Nizar Bouguila</name><affiliation wicri:level="1"><country xml:lang="fr">Canada</country><wicri:regionArea>Concordia Institute for Information Systems Engineering (CIISE), Concordia University, H3G 1T7, Montreal, QC</wicri:regionArea><wicri:noRegion>QC</wicri:noRegion></affiliation><affiliation wicri:level="1"><country wicri:rule="url">Canada</country></affiliation></author><author><name sortKey="Ziou, Djemel" sort="Ziou, Djemel" uniqKey="Ziou D" first="Djemel" last="Ziou">Djemel Ziou</name><affiliation wicri:level="1"><country xml:lang="fr">Canada</country><wicri:regionArea>Département d’Informatique, Université de Sherbrooke, J1K 2R1, Sherbrooke, QC</wicri:regionArea><wicri:noRegion>QC</wicri:noRegion></affiliation><affiliation wicri:level="1"><country wicri:rule="url">Canada</country></affiliation></author><author><name sortKey="Hammoud, Riad I" sort="Hammoud, Riad I" uniqKey="Hammoud R" first="Riad I." last="Hammoud">Riad I. Hammoud</name><affiliation wicri:level="2"><country xml:lang="fr">États-Unis</country><wicri:regionArea>Delphi Corporation, Delphi Electronics and Safety, Kokomo, IN</wicri:regionArea><placeName><region type="state">Indiana</region></placeName></affiliation><affiliation></affiliation></author></analytic><monogr></monogr><series><title level="j">Pattern Analysis and Applications</title><title level="j" type="abbrev">Pattern Anal Applic</title><idno type="ISSN">1433-7541</idno><idno type="eISSN">1433-755X</idno><imprint><publisher>Springer-Verlag</publisher><pubPlace>London</pubPlace><date type="published" when="2009-06-01">2009-06-01</date><biblScope unit="volume">12</biblScope><biblScope unit="issue">2</biblScope><biblScope unit="page" from="151">151</biblScope><biblScope unit="page" to="166">166</biblScope></imprint><idno type="ISSN">1433-7541</idno></series></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">1433-7541</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Bayesian analysis</term><term>EM</term><term>Generalized Dirichlet mixture</term><term>Gibbs sampling</term><term>Images classification</term><term>Metropolis–Hastings</term><term>SVM</term></keywords></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Abstract: In this paper, we present a fully Bayesian approach for generalized Dirichlet mixtures estimation and selection. The estimation of the parameters is based on the Monte Carlo simulation technique of Gibbs sampling mixed with a Metropolis-Hastings step. Also, we obtain a posterior distribution which is conjugate to a generalized Dirichlet likelihood. For the selection of the number of clusters, we used the integrated likelihood. The performance of our Bayesian algorithm is tested and compared with the maximum likelihood approach by the classification of several synthetic and real data sets. The generalized Dirichlet mixture is also applied to the problems of IR eye modeling and introduced as a probabilistic kernel for Support Vector Machines.</div></front></TEI><affiliations><list><country><li>Canada</li><li>États-Unis</li></country><region><li>Indiana</li></region></list><tree><country name="Canada"><noRegion><name sortKey="Bouguila, Nizar" sort="Bouguila, Nizar" uniqKey="Bouguila N" first="Nizar" last="Bouguila">Nizar Bouguila</name></noRegion><name sortKey="Bouguila, Nizar" sort="Bouguila, Nizar" uniqKey="Bouguila N" first="Nizar" last="Bouguila">Nizar Bouguila</name><name sortKey="Ziou, Djemel" sort="Ziou, Djemel" uniqKey="Ziou D" first="Djemel" last="Ziou">Djemel Ziou</name><name sortKey="Ziou, Djemel" sort="Ziou, Djemel" uniqKey="Ziou D" first="Djemel" last="Ziou">Djemel Ziou</name></country><country name="États-Unis"><region name="Indiana"><name sortKey="Hammoud, Riad I" sort="Hammoud, Riad I" uniqKey="Hammoud R" first="Riad I." last="Hammoud">Riad I. Hammoud</name></region></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 004345 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 004345 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Lorraine
|area= InforLorV4
|flux= Main
|étape= Exploration
|type= RBID
|clé= ISTEX:D65DF631C50E3E2230F33877937834B44BE7F510
|texte= On Bayesian analysis of a finite generalized Dirichlet mixture via a Metropolis-within-Gibbs sampling
}}
| This area was generated with Dilib version V0.6.33. |  |