Active Learning in Parallel Universes

Cebron, Nicolas; Berthold, Michael R.

Active Learning in Parallel Universes

Dateien

Cebron_243122.pdfGröße: 190.95 KBDownloads: 171

Datum

2010

Autor:innen

Cebron, Nicolas

Berthold, Michael R.

Open Access-Veröffentlichung

Open Access Green

Sammlungen

Informatik und Informationswissenschaft

Publikationstyp

Beitrag zu einem Konferenzband

Publikationsstatus

Published

Erschienen in

Proceedings of the 19th ACM international conference on Information and knowledge management - CIKM '10. New York, New York, USA: ACM Press, 2010, pp. 1621-1624. ISBN 978-1-4503-0099-5. Available under: doi: 10.1145/1871437.1871688

Zusammenfassung

This work addresses two challenges in combination: learning with a very limited number of labeled training examples (active learning) and learning in the presence of multiple views for each object where the global model to be learned is spread out over some or all of these views (learning in parallel universes). We propose a new active learning approach which selects the best samples to query the label with the goal of improving overall model accuracy and determining which universe contributes most to the local model. The resulting combination and class-specific weighting of universes provides a significantly better classification accuracy than traditional active learning methods.

Fachgebiet (DDC)

004 Informatik

Konferenz

the 19th ACM international conference, 26. Okt. 2010 - 30. Okt. 2010, Toronto, ON, Canada

Zitieren

ISO 690

CEBRON, Nicolas, Michael R. BERTHOLD, 2010. Active Learning in Parallel Universes. the 19th ACM international conference. Toronto, ON, Canada, 26. Okt. 2010 - 30. Okt. 2010. In: Proceedings of the 19th ACM international conference on Information and knowledge management - CIKM '10. New York, New York, USA: ACM Press, 2010, pp. 1621-1624. ISBN 978-1-4503-0099-5. Available under: doi: 10.1145/1871437.1871688

BibTex

@inproceedings{Cebron2010Activ-24312,
  year={2010},
  doi={10.1145/1871437.1871688},
  title={Active Learning in Parallel Universes},
  isbn={978-1-4503-0099-5},
  publisher={ACM Press},
  address={New York, New York, USA},
  booktitle={Proceedings of the 19th ACM international conference on Information and knowledge management - CIKM '10},
  pages={1621--1624},
  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/123456789/24312">
    <dcterms:issued>2010</dcterms:issued>
    <dcterms:title>Active Learning in Parallel Universes</dcterms:title>
    <dcterms:bibliographicCitation>CIKM'10 : proceedings of the 19th International Conference on Information &amp; Knowledge Management and Co-Located Workshops; Oktober 26 - 30, 2010, Toronto, Ontario, Canada / ed. by Xiangji Jimmy Huang .... - New York, N.Y. : ACM, Association for Computing Machinery, 2010. - S. 1621-1624. - ISBN 978-1-4503-0099-5</dcterms:bibliographicCitation>
    <dcterms:hasPart rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/24312/2/Cebron_243122.pdf"/>
    <dc:contributor>Berthold, Michael R.</dc:contributor>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dspace:hasBitstream rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/24312/2/Cebron_243122.pdf"/>
    <dc:language>eng</dc:language>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2013-08-23T12:28:44Z</dc:date>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
    <dcterms:rights rdf:resource="https://rightsstatements.org/page/InC/1.0/"/>
    <dc:creator>Berthold, Michael R.</dc:creator>
    <dcterms:abstract xml:lang="eng">This work addresses two challenges in combination: learning with a very limited number of labeled training examples (active learning) and learning in the presence of multiple views for each object where the global model to be learned is spread out over some or all of these views (learning in parallel universes). We propose a new active learning approach which selects the best samples to query the label with the goal of improving overall model accuracy and determining which universe contributes most to the local model. The resulting combination and class-specific weighting of universes provides a significantly better classification accuracy than traditional active learning methods.</dcterms:abstract>
    <dc:creator>Cebron, Nicolas</dc:creator>
    <dc:rights>terms-of-use</dc:rights>
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <bibo:uri rdf:resource="http://kops.uni-konstanz.de/handle/123456789/24312"/>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2013-08-23T12:28:44Z</dcterms:available>
    <dc:contributor>Cebron, Nicolas</dc:contributor>
  </rdf:Description>
</rdf:RDF>

Universitätsbibliographie

Ja