Reducing the Effect of Out-Voting Problem in Ensemble Based Incremental Support Vector Machines
Identifieur interne : 001256 ( Main/Exploration ); précédent : 001255; suivant : 001257Reducing the Effect of Out-Voting Problem in Ensemble Based Incremental Support Vector Machines
Auteurs : Zeki Erdem [Turquie] ; Robi Polikar [États-Unis] ; Fikret Gurgen [Turquie] ; Nejat Yumusak [Turquie]Source :
- Lecture Notes in Computer Science [ 0302-9743 ] ; 2005.
Abstract
Abstract: Although Support Vector Machines (SVMs) have been successfully applied to solve a large number of classification and regression problems, they suffer from the catastrophic forgetting phenomenon. In our previous work, integrating the SVM classifiers into an ensemble framework using Learn++ (SVMLearn++) [1], we have shown that the SVM classifiers can in fact be equipped with the incremental learning capability. However, Learn++ suffers from an inherent out-voting problem: when asked to learn new classes, an unnecessarily large number of classifiers are generated to learn the new classes. In this paper, we propose a new ensemble based incremental learning approach using SVMs that is based on the incremental Learn++.MT algorithm. Experiments on the real-world and benchmark datasets show that the proposed approach can reduce the number of SVM classifiers generated, thus reduces the effect of out-voting problem. It also provides performance improvements over previous approach.
Url:
DOI: 10.1007/11550907_96
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 000E31
- to stream Istex, to step Curation: 000D98
- to stream Istex, to step Checkpoint: 000B65
- to stream Main, to step Merge: 001292
- to stream Main, to step Curation: 001256
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Reducing the Effect of Out-Voting Problem in Ensemble Based Incremental Support Vector Machines</title><author><name sortKey="Erdem, Zeki" sort="Erdem, Zeki" uniqKey="Erdem Z" first="Zeki" last="Erdem">Zeki Erdem</name></author><author><name sortKey="Polikar, Robi" sort="Polikar, Robi" uniqKey="Polikar R" first="Robi" last="Polikar">Robi Polikar</name></author><author><name sortKey="Gurgen, Fikret" sort="Gurgen, Fikret" uniqKey="Gurgen F" first="Fikret" last="Gurgen">Fikret Gurgen</name></author><author><name sortKey="Yumusak, Nejat" sort="Yumusak, Nejat" uniqKey="Yumusak N" first="Nejat" last="Yumusak">Nejat Yumusak</name></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:7155B31BC88BB2678B987C67811112604D551617</idno><date when="2005" year="2005">2005</date><idno type="doi">10.1007/11550907_96</idno><idno type="url">https://api.istex.fr/document/7155B31BC88BB2678B987C67811112604D551617/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">000E31</idno><idno type="wicri:Area/Istex/Curation">000D98</idno><idno type="wicri:Area/Istex/Checkpoint">000B65</idno><idno type="wicri:doubleKey">0302-9743:2005:Erdem Z:reducing:the:effect</idno><idno type="wicri:Area/Main/Merge">001292</idno><idno type="wicri:Area/Main/Curation">001256</idno><idno type="wicri:Area/Main/Exploration">001256</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Reducing the Effect of Out-Voting Problem in Ensemble Based Incremental Support Vector Machines</title><author><name sortKey="Erdem, Zeki" sort="Erdem, Zeki" uniqKey="Erdem Z" first="Zeki" last="Erdem">Zeki Erdem</name><affiliation wicri:level="1"><country xml:lang="fr">Turquie</country><wicri:regionArea>Information Technologies Institute, TUBITAK Marmara Research Center, 41470, Gebze – Kocaeli</wicri:regionArea><wicri:noRegion>Gebze – Kocaeli</wicri:noRegion></affiliation><affiliation wicri:level="1"><country xml:lang="fr">Turquie</country><wicri:regionArea>Computer Engineering Department, Sakarya University, 54187, Esentepe, Sakarya</wicri:regionArea><wicri:noRegion>Sakarya</wicri:noRegion></affiliation><affiliation wicri:level="1"><country wicri:rule="url">Turquie</country></affiliation></author><author><name sortKey="Polikar, Robi" sort="Polikar, Robi" uniqKey="Polikar R" first="Robi" last="Polikar">Robi Polikar</name><affiliation wicri:level="2"><country xml:lang="fr">États-Unis</country><wicri:regionArea>Electrical and Computer Engineering Department, Rowan University, 210 Mullica Hill Rd., 08028, Glassboro, NJ</wicri:regionArea><placeName><region type="state">New Jersey</region></placeName></affiliation><affiliation wicri:level="1"><country wicri:rule="url">États-Unis</country></affiliation></author><author><name sortKey="Gurgen, Fikret" sort="Gurgen, Fikret" uniqKey="Gurgen F" first="Fikret" last="Gurgen">Fikret Gurgen</name><affiliation wicri:level="1"><country xml:lang="fr">Turquie</country><wicri:regionArea>Computer Engineering Department, Bogazici University, 80815, Bebek, Istanbul</wicri:regionArea><wicri:noRegion>Istanbul</wicri:noRegion></affiliation><affiliation wicri:level="1"><country wicri:rule="url">Turquie</country></affiliation></author><author><name sortKey="Yumusak, Nejat" sort="Yumusak, Nejat" uniqKey="Yumusak N" first="Nejat" last="Yumusak">Nejat Yumusak</name><affiliation wicri:level="1"><country xml:lang="fr">Turquie</country><wicri:regionArea>Computer Engineering Department, Sakarya University, 54187, Esentepe, Sakarya</wicri:regionArea><wicri:noRegion>Sakarya</wicri:noRegion></affiliation><affiliation wicri:level="1"><country wicri:rule="url">Turquie</country></affiliation></author></analytic><monogr></monogr><series><title level="s">Lecture Notes in Computer Science</title><imprint><date>2005</date></imprint><idno type="ISSN">0302-9743</idno><idno type="eISSN">1611-3349</idno></series><idno type="istex">7155B31BC88BB2678B987C67811112604D551617</idno><idno type="DOI">10.1007/11550907_96</idno><idno type="ChapterID">96</idno><idno type="ChapterID">Chap96</idno></biblStruct></sourceDesc><seriesStmt><idno type="doi">10.1007/11550907_163</idno></seriesStmt></fileDesc><profileDesc><textClass></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Abstract: Although Support Vector Machines (SVMs) have been successfully applied to solve a large number of classification and regression problems, they suffer from the catastrophic forgetting phenomenon. In our previous work, integrating the SVM classifiers into an ensemble framework using Learn++ (SVMLearn++) [1], we have shown that the SVM classifiers can in fact be equipped with the incremental learning capability. However, Learn++ suffers from an inherent out-voting problem: when asked to learn new classes, an unnecessarily large number of classifiers are generated to learn the new classes. In this paper, we propose a new ensemble based incremental learning approach using SVMs that is based on the incremental Learn++.MT algorithm. Experiments on the real-world and benchmark datasets show that the proposed approach can reduce the number of SVM classifiers generated, thus reduces the effect of out-voting problem. It also provides performance improvements over previous approach.</div></front></TEI><affiliations><list><country><li>Turquie</li><li>États-Unis</li></country><region><li>New Jersey</li></region></list><tree><country name="Turquie"><noRegion><name sortKey="Erdem, Zeki" sort="Erdem, Zeki" uniqKey="Erdem Z" first="Zeki" last="Erdem">Zeki Erdem</name></noRegion><name sortKey="Erdem, Zeki" sort="Erdem, Zeki" uniqKey="Erdem Z" first="Zeki" last="Erdem">Zeki Erdem</name><name sortKey="Erdem, Zeki" sort="Erdem, Zeki" uniqKey="Erdem Z" first="Zeki" last="Erdem">Zeki Erdem</name><name sortKey="Gurgen, Fikret" sort="Gurgen, Fikret" uniqKey="Gurgen F" first="Fikret" last="Gurgen">Fikret Gurgen</name><name sortKey="Gurgen, Fikret" sort="Gurgen, Fikret" uniqKey="Gurgen F" first="Fikret" last="Gurgen">Fikret Gurgen</name><name sortKey="Yumusak, Nejat" sort="Yumusak, Nejat" uniqKey="Yumusak N" first="Nejat" last="Yumusak">Nejat Yumusak</name><name sortKey="Yumusak, Nejat" sort="Yumusak, Nejat" uniqKey="Yumusak N" first="Nejat" last="Yumusak">Nejat Yumusak</name></country><country name="États-Unis"><region name="New Jersey"><name sortKey="Polikar, Robi" sort="Polikar, Robi" uniqKey="Polikar R" first="Robi" last="Polikar">Robi Polikar</name></region><name sortKey="Polikar, Robi" sort="Polikar, Robi" uniqKey="Polikar R" first="Robi" last="Polikar">Robi Polikar</name></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001256 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 001256 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= OcrV1
|flux= Main
|étape= Exploration
|type= RBID
|clé= ISTEX:7155B31BC88BB2678B987C67811112604D551617
|texte= Reducing the Effect of Out-Voting Problem in Ensemble Based Incremental Support Vector Machines
}}
| This area was generated with Dilib version V0.6.32. |  |