Architecture for Classifier Combination Using Entropy Measures
Identifieur interne : 001D88 ( Main/Exploration ); précédent : 001D87; suivant : 001D89Architecture for Classifier Combination Using Entropy Measures
Auteurs : Kr. Ianakiev [États-Unis] ; Venugopal Govindaraju [États-Unis]Source :
- Lecture Notes in Computer Science [ 0302-9743 ] ; 2000.
Abstract
Abstract: In this paper we emphasize the need for a general theory of combination. Presently, most systems combine recognizers in an ad hoc manner. Recognizers can be combined in series and/or in parallel. Empirical methods can become extremely time consuming, given the very large number of combination possibilities. We have developed a method of systematically arriving at the optimal architecture for combination of classifiers that can include both parallel and serial methods. Our focus in this paper, however, will be on serial methods. We also derive some theoretical results to lay the foundation for our experiments. We show how a greedy algorithm that strives for entropy reduction at every stage leads to results superior to combination methods which are ad hoc. In our experiments we have seen an advantage of about 5% in certain cases.
Url:
DOI: 10.1007/3-540-45014-9_33
Affiliations:
- États-Unis
- État de New York
- Buffalo (New York)
- Université d'État de New York, Université d'État de New York à Buffalo
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 003245
- to stream Istex, to step Curation: 003000
- to stream Istex, to step Checkpoint: 001391
- to stream Main, to step Merge: 001E88
- to stream Main, to step Curation: 001D88
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Architecture for Classifier Combination Using Entropy Measures</title>
<author><name sortKey="Ianakiev, Kr" sort="Ianakiev, Kr" uniqKey="Ianakiev K" first="Kr." last="Ianakiev">Kr. Ianakiev</name>
</author>
<author><name sortKey="Govindaraju, V" sort="Govindaraju, V" uniqKey="Govindaraju V" first="V." last="Govindaraju">Venugopal Govindaraju</name>
<affiliation><country>États-Unis</country>
<placeName><settlement type="city">Buffalo (New York)</settlement>
<region type="state">État de New York</region>
</placeName>
<orgName type="university" n="3">Université d'État de New York à Buffalo</orgName>
<orgName type="institution">Université d'État de New York</orgName>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:E3F9F45038250820605213CD895EAD921D62F0C6</idno>
<date when="2000" year="2000">2000</date>
<idno type="doi">10.1007/3-540-45014-9_33</idno>
<idno type="url">https://api.istex.fr/document/E3F9F45038250820605213CD895EAD921D62F0C6/fulltext/pdf</idno>
<idno type="wicri:Area/Istex/Corpus">003245</idno>
<idno type="wicri:Area/Istex/Curation">003000</idno>
<idno type="wicri:Area/Istex/Checkpoint">001391</idno>
<idno type="wicri:doubleKey">0302-9743:2000:Ianakiev K:architecture:for:classifier</idno>
<idno type="wicri:Area/Main/Merge">001E88</idno>
<idno type="wicri:Area/Main/Curation">001D88</idno>
<idno type="wicri:Area/Main/Exploration">001D88</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Architecture for Classifier Combination Using Entropy Measures</title>
<author><name sortKey="Ianakiev, Kr" sort="Ianakiev, Kr" uniqKey="Ianakiev K" first="Kr." last="Ianakiev">Kr. Ianakiev</name>
<affiliation wicri:level="4"><country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Center for Excellence in Document Analysis and Recognition (CEDAR) Department of Computer Science & Engineering, University at Buffalo, 14228, Amherst, NY</wicri:regionArea>
<placeName><region type="state">État de New York</region>
<settlement type="city">Buffalo (New York)</settlement>
</placeName>
<orgName type="university">Université d'État de New York à Buffalo</orgName>
</affiliation>
<affiliation wicri:level="1"><country wicri:rule="url">États-Unis</country>
</affiliation>
</author>
<author><name sortKey="Govindaraju, V" sort="Govindaraju, V" uniqKey="Govindaraju V" first="V." last="Govindaraju">Venugopal Govindaraju</name>
<affiliation wicri:level="4"><country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Center for Excellence in Document Analysis and Recognition (CEDAR) Department of Computer Science & Engineering, University at Buffalo, 14228, Amherst, NY</wicri:regionArea>
<placeName><region type="state">État de New York</region>
<settlement type="city">Buffalo (New York)</settlement>
</placeName>
<orgName type="university">Université d'État de New York à Buffalo</orgName>
<placeName><settlement type="city">Buffalo (New York)</settlement>
<region type="state">État de New York</region>
</placeName>
<orgName type="university" n="3">Université d'État de New York à Buffalo</orgName>
<orgName type="institution">Université d'État de New York</orgName>
</affiliation>
<affiliation wicri:level="1"><country wicri:rule="url">États-Unis</country>
<placeName><settlement type="city">Buffalo (New York)</settlement>
<region type="state">État de New York</region>
</placeName>
<orgName type="university" n="3">Université d'État de New York à Buffalo</orgName>
<orgName type="institution">Université d'État de New York</orgName>
</affiliation>
</author>
</analytic>
<monogr></monogr>
<series><title level="s">Lecture Notes in Computer Science</title>
<imprint><date>2000</date>
</imprint>
<idno type="ISSN">0302-9743</idno>
<idno type="ISSN">0302-9743</idno>
</series>
<idno type="istex">E3F9F45038250820605213CD895EAD921D62F0C6</idno>
<idno type="DOI">10.1007/3-540-45014-9_33</idno>
<idno type="ChapterID">33</idno>
<idno type="ChapterID">Chap33</idno>
</biblStruct>
</sourceDesc>
<seriesStmt><idno type="ISSN">0302-9743</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass></textClass>
<langUsage><language ident="en">en</language>
</langUsage>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Abstract: In this paper we emphasize the need for a general theory of combination. Presently, most systems combine recognizers in an ad hoc manner. Recognizers can be combined in series and/or in parallel. Empirical methods can become extremely time consuming, given the very large number of combination possibilities. We have developed a method of systematically arriving at the optimal architecture for combination of classifiers that can include both parallel and serial methods. Our focus in this paper, however, will be on serial methods. We also derive some theoretical results to lay the foundation for our experiments. We show how a greedy algorithm that strives for entropy reduction at every stage leads to results superior to combination methods which are ad hoc. In our experiments we have seen an advantage of about 5% in certain cases.</div>
</front>
</TEI>
<affiliations><list><country><li>États-Unis</li>
</country>
<region><li>État de New York</li>
</region>
<settlement><li>Buffalo (New York)</li>
</settlement>
<orgName><li>Université d'État de New York</li>
<li>Université d'État de New York à Buffalo</li>
</orgName>
</list>
<tree><country name="États-Unis"><region name="État de New York"><name sortKey="Ianakiev, Kr" sort="Ianakiev, Kr" uniqKey="Ianakiev K" first="Kr." last="Ianakiev">Kr. Ianakiev</name>
</region>
<name sortKey="Govindaraju, V" sort="Govindaraju, V" uniqKey="Govindaraju V" first="V." last="Govindaraju">Venugopal Govindaraju</name>
<name sortKey="Govindaraju, V" sort="Govindaraju, V" uniqKey="Govindaraju V" first="V." last="Govindaraju">Venugopal Govindaraju</name>
<name sortKey="Ianakiev, Kr" sort="Ianakiev, Kr" uniqKey="Ianakiev K" first="Kr." last="Ianakiev">Kr. Ianakiev</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 001D88 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 001D88 | 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:E3F9F45038250820605213CD895EAD921D62F0C6 |texte= Architecture for Classifier Combination Using Entropy Measures }}
This area was generated with Dilib version V0.6.32. |