Active Learning for Object Classification : From Exploration to Exploitation

Cebron, Nicolas; Berthold, Michael R.

Active Learning for Object Classification : From Exploration to Exploitation

Dateien

Active Learning for Object-erl.pdfGröße: 6.92 MBDownloads: 786

Datum

2008

Autor:innen

Cebron, Nicolas

Berthold, Michael R.

Open Access-Veröffentlichung

Open Access Green

Sammlungen

Informatik und Informationswissenschaft

Publikationstyp

Zeitschriftenartikel

Publikationsstatus

Published

Erschienen in

Data Mining and Knowledge Discovery. 2008, 18(2), pp. 283-299. ISSN 1384-5810. eISSN 1573-756X. Available under: doi: 10.1007/s10618-008-0115-0

Zusammenfassung

Classifying large datasets without any a-priori information poses a problem in numerous tasks. Especially in industrial environments, we often encounter diverse measurement devices and sensors that produce huge amounts of data, but we still rely on a human expert to help give the data a meaningful interpretation. As the amount of data that must be manually classified plays a critical role, we need to reduce the number of learning episodes involving human interactions as much as possible. In addition for real world applications it is fundamental to converge in a stable manner to a solution that is close to the optimal solution. We present a new self-controlled exploration/exploitation strategy to select data points to be labeled by a domain expert where the potential of each data point is computed based on a combination of its representativeness and the uncertainty of the classifier. A new Prototype Based Active Learning (PBAC) algorithm for classification is introduced. We compare the results to other active learning approaches on several benchmark datasets.

Fachgebiet (DDC)

004 Informatik

Schlagwörter

active learning, data mining, subtractive clustering, exploration, exploitation, prototype classification

Zitieren

ISO 690

CEBRON, Nicolas, Michael R. BERTHOLD, 2008. Active Learning for Object Classification : From Exploration to Exploitation. In: Data Mining and Knowledge Discovery. 2008, 18(2), pp. 283-299. ISSN 1384-5810. eISSN 1573-756X. Available under: doi: 10.1007/s10618-008-0115-0

BibTex

@article{Cebron2008Activ-3028,
  year={2008},
  doi={10.1007/s10618-008-0115-0},
  title={Active Learning for Object Classification : From Exploration to Exploitation},
  number={2},
  volume={18},
  issn={1384-5810},
  journal={Data Mining and Knowledge Discovery},
  pages={283--299},
  author={Cebron, Nicolas and Berthold, Michael R.}
}

RDF

<rdf:RDF
    xmlns:dcterms="http://purl.org/dc/terms/"
    xmlns:dc="http://purl.org/dc/elements/1.1/"
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:bibo="http://purl.org/ontology/bibo/"
    xmlns:dspace="http://digital-repositories.org/ontologies/dspace/0.1.0#"
    xmlns:foaf="http://xmlns.com/foaf/0.1/"
    xmlns:void="http://rdfs.org/ns/void#"
    xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > 
  <rdf:Description rdf:about="https://kops.uni-konstanz.de/server/rdf/resource/123456789/3028">
    <bibo:uri rdf:resource="http://kops.uni-konstanz.de/handle/123456789/3028"/>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2011-03-23T10:15:50Z</dcterms:available>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2011-03-23T10:15:50Z</dc:date>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:creator>Cebron, Nicolas</dc:creator>
    <dcterms:issued>2008</dcterms:issued>
    <dspace:hasBitstream rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/3028/1/Active%20Learning%20for%20Object-erl.pdf"/>
    <dcterms:title>Active Learning for Object Classification : From Exploration to Exploitation</dcterms:title>
    <dcterms:bibliographicCitation>Data Mining and Knowledge Discovery ; 18 (2009), 2. - S. 283-299</dcterms:bibliographicCitation>
    <dcterms:rights rdf:resource="https://rightsstatements.org/page/InC/1.0/"/>
    <dcterms:hasPart rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/3028/1/Active%20Learning%20for%20Object-erl.pdf"/>
    <dc:creator>Berthold, Michael R.</dc:creator>
    <dc:contributor>Cebron, Nicolas</dc:contributor>
    <dc:contributor>Berthold, Michael R.</dc:contributor>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
    <dc:rights>terms-of-use</dc:rights>
    <dcterms:abstract xml:lang="eng">Classifying large datasets without any a-priori information poses a problem in numerous tasks. Especially in industrial environments, we often encounter diverse measurement devices and sensors that produce huge amounts of data, but we still rely on a human expert to help give the data a meaningful interpretation. As the amount of data that must be manually classified plays a critical role, we need to reduce the number of learning episodes involving human interactions as much as possible. In addition for real world applications it is fundamental to converge in a stable manner to a solution that is close to the optimal solution. We present a new self-controlled exploration/exploitation strategy to select data points to be labeled by a domain expert where the potential of each data point is computed based on a combination of its representativeness and the uncertainty of the classifier. A new Prototype Based Active Learning (PBAC) algorithm for classification is introduced. We compare the results to other active learning approaches on several benchmark datasets.</dcterms:abstract>
    <dc:language>eng</dc:language>
  </rdf:Description>
</rdf:RDF>

Universitätsbibliographie

Ja