Optimal Classifier Fusion in a Non-Bayesian Probabilistic Framework
Identifieur interne : 003B71 ( Main/Exploration ); précédent : 003B70; suivant : 003B72Optimal Classifier Fusion in a Non-Bayesian Probabilistic Framework
Auteurs : Oriol Ramos Terrades [Espagne] ; Ernest Valveny [Espagne] ; Salvatore Tabbone [France]Source :
- IEEE transactions on pattern analysis and machine intelligence [ 0162-8828 ] ; 2009.
Descripteurs français
- Pascal (Inist)
- Wicri :
- topic : Classification, Base de données.
English descriptors
- KwdEn :
Abstract
The combination of the output of classifiers has been one of the strategies used to improve classification rates in general purpose classification systems. Some of the most common approaches can be explained using the Bayes' formula. In this paper, we tackle the problem of the combination of classifiers using a non-Bayesian probabilistic framework. This approach permits us to derive two linear combination rules that minimize misclassification rates under some constraints on the distribution of classifiers. In order to show the validity of this approach we have compared it with other popular combination rules from a theoretical viewpoint using a synthetic data set, and experimentally using two standard databases: the MNIST handwritten digit database and the GREC symbol database. Results on the synthetic data set show the validity of the theoretical approach. Indeed, results on real data show that the proposed methods outperform other common combination schemes.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000234
- to stream PascalFrancis, to step Curation: 000784
- to stream PascalFrancis, to step Checkpoint: 000233
- to stream Main, to step Merge: 003C67
- to stream Main, to step Curation: 003B71
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Optimal Classifier Fusion in a Non-Bayesian Probabilistic Framework</title><author><name sortKey="Terrades, Oriol Ramos" sort="Terrades, Oriol Ramos" uniqKey="Terrades O" first="Oriol Ramos" last="Terrades">Oriol Ramos Terrades</name><affiliation wicri:level="2"><inist:fA14 i1="01"><s1>Computer Vision Centre and the Department of Computer Science, Universitat Automous of Barcelona, Edifici O, Campus UAB</s1><s2>08193 Bellaterra</s2><s3>ESP</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ></inist:fA14><country>Espagne</country><placeName><region nuts="2" type="communauté">Catalogne</region></placeName></affiliation></author><author><name sortKey="Valveny, Ernest" sort="Valveny, Ernest" uniqKey="Valveny E" first="Ernest" last="Valveny">Ernest Valveny</name><affiliation wicri:level="2"><inist:fA14 i1="01"><s1>Computer Vision Centre and the Department of Computer Science, Universitat Automous of Barcelona, Edifici O, Campus UAB</s1><s2>08193 Bellaterra</s2><s3>ESP</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ></inist:fA14><country>Espagne</country><placeName><region nuts="2" type="communauté">Catalogne</region></placeName></affiliation></author><author><name sortKey="Tabbone, Salvatore" sort="Tabbone, Salvatore" uniqKey="Tabbone S" first="Salvatore" last="Tabbone">Salvatore Tabbone</name><affiliation wicri:level="3"><inist:fA14 i1="02"><s1>LORIA and the University of Nancy 2, Campus Scientifique, BP 239</s1><s2>54506 Vandœvre-les-Nancy</s2><s3>FRA</s3><sZ>3 aut.</sZ></inist:fA14><country>France</country><placeName><region type="region" nuts="2">Grand Est</region><region type="old region" nuts="2">Lorraine (région)</region><settlement type="city">Vandœvre-les-Nancy</settlement></placeName></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">INIST</idno><idno type="inist">10-0070511</idno><date when="2009">2009</date><idno type="stanalyst">PASCAL 10-0070511 INIST</idno><idno type="RBID">Pascal:10-0070511</idno><idno type="wicri:Area/PascalFrancis/Corpus">000234</idno><idno type="wicri:Area/PascalFrancis/Curation">000784</idno><idno type="wicri:Area/PascalFrancis/Checkpoint">000233</idno><idno type="wicri:explorRef" wicri:stream="PascalFrancis" wicri:step="Checkpoint">000233</idno><idno type="wicri:doubleKey">0162-8828:2009:Terrades O:optimal:classifier:fusion</idno><idno type="wicri:Area/Main/Merge">003C67</idno><idno type="wicri:Area/Main/Curation">003B71</idno><idno type="wicri:Area/Main/Exploration">003B71</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Optimal Classifier Fusion in a Non-Bayesian Probabilistic Framework</title><author><name sortKey="Terrades, Oriol Ramos" sort="Terrades, Oriol Ramos" uniqKey="Terrades O" first="Oriol Ramos" last="Terrades">Oriol Ramos Terrades</name><affiliation wicri:level="2"><inist:fA14 i1="01"><s1>Computer Vision Centre and the Department of Computer Science, Universitat Automous of Barcelona, Edifici O, Campus UAB</s1><s2>08193 Bellaterra</s2><s3>ESP</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ></inist:fA14><country>Espagne</country><placeName><region nuts="2" type="communauté">Catalogne</region></placeName></affiliation></author><author><name sortKey="Valveny, Ernest" sort="Valveny, Ernest" uniqKey="Valveny E" first="Ernest" last="Valveny">Ernest Valveny</name><affiliation wicri:level="2"><inist:fA14 i1="01"><s1>Computer Vision Centre and the Department of Computer Science, Universitat Automous of Barcelona, Edifici O, Campus UAB</s1><s2>08193 Bellaterra</s2><s3>ESP</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ></inist:fA14><country>Espagne</country><placeName><region nuts="2" type="communauté">Catalogne</region></placeName></affiliation></author><author><name sortKey="Tabbone, Salvatore" sort="Tabbone, Salvatore" uniqKey="Tabbone S" first="Salvatore" last="Tabbone">Salvatore Tabbone</name><affiliation wicri:level="3"><inist:fA14 i1="02"><s1>LORIA and the University of Nancy 2, Campus Scientifique, BP 239</s1><s2>54506 Vandœvre-les-Nancy</s2><s3>FRA</s3><sZ>3 aut.</sZ></inist:fA14><country>France</country><placeName><region type="region" nuts="2">Grand Est</region><region type="old region" nuts="2">Lorraine (région)</region><settlement type="city">Vandœvre-les-Nancy</settlement></placeName></affiliation></author></analytic><series><title level="j" type="main">IEEE transactions on pattern analysis and machine intelligence</title><title level="j" type="abbreviated">IEEE trans. pattern anal. mach. intell.</title><idno type="ISSN">0162-8828</idno><imprint><date when="2009">2009</date></imprint></series></biblStruct></sourceDesc><seriesStmt><title level="j" type="main">IEEE transactions on pattern analysis and machine intelligence</title><title level="j" type="abbreviated">IEEE trans. pattern anal. mach. intell.</title><idno type="ISSN">0162-8828</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Bayes estimation</term><term>Classification</term><term>Data fusion</term><term>Database</term><term>Manuscript character</term><term>Modeling</term><term>Probabilistic approach</term><term>Random variable</term></keywords><keywords scheme="Pascal" xml:lang="fr"><term>Fusion donnée</term><term>Classification</term><term>Base de données</term><term>Caractère manuscrit</term><term>Estimation Bayes</term><term>Approche probabiliste</term><term>Modélisation</term><term>Variable aléatoire</term><term>.</term></keywords><keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Classification</term><term>Base de données</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">The combination of the output of classifiers has been one of the strategies used to improve classification rates in general purpose classification systems. Some of the most common approaches can be explained using the Bayes' formula. In this paper, we tackle the problem of the combination of classifiers using a non-Bayesian probabilistic framework. This approach permits us to derive two linear combination rules that minimize misclassification rates under some constraints on the distribution of classifiers. In order to show the validity of this approach we have compared it with other popular combination rules from a theoretical viewpoint using a synthetic data set, and experimentally using two standard databases: the MNIST handwritten digit database and the GREC symbol database. Results on the synthetic data set show the validity of the theoretical approach. Indeed, results on real data show that the proposed methods outperform other common combination schemes.</div></front></TEI><affiliations><list><country><li>Espagne</li><li>France</li></country><region><li>Catalogne</li><li>Grand Est</li><li>Lorraine (région)</li></region><settlement><li>Vandœvre-les-Nancy</li></settlement></list><tree><country name="Espagne"><region name="Catalogne"><name sortKey="Terrades, Oriol Ramos" sort="Terrades, Oriol Ramos" uniqKey="Terrades O" first="Oriol Ramos" last="Terrades">Oriol Ramos Terrades</name></region><name sortKey="Valveny, Ernest" sort="Valveny, Ernest" uniqKey="Valveny E" first="Ernest" last="Valveny">Ernest Valveny</name></country><country name="France"><region name="Grand Est"><name sortKey="Tabbone, Salvatore" sort="Tabbone, Salvatore" uniqKey="Tabbone S" first="Salvatore" last="Tabbone">Salvatore Tabbone</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 003B71 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 003B71 | 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= Pascal:10-0070511
|texte= Optimal Classifier Fusion in a Non-Bayesian Probabilistic Framework
}}
| This area was generated with Dilib version V0.6.33. |  |