An Adaptive Multi Objective Selection Strategy for Active Learning


Dateien zu dieser Ressource

Prüfsumme: MD5:b78f0b10e52b344363923e0b0116d416

CEBRON, Nicolas, Michael R. BERTHOLD, 2007. An Adaptive Multi Objective Selection Strategy for Active Learning

@techreport{Cebron2007Adapt-6148, series={Konstanzer Schriften in Mathematik und Informatik}, title={An Adaptive Multi Objective Selection Strategy for Active Learning}, year={2007}, number={235}, author={Cebron, Nicolas and Berthold, Michael R.} }

<rdf:RDF xmlns:dcterms="" xmlns:dc="" xmlns:rdf="" xmlns:bibo="" xmlns:dspace="" xmlns:foaf="" xmlns:void="" xmlns:xsd="" > <rdf:Description rdf:about=""> <foaf:homepage rdf:resource="http://localhost:8080/jspui"/> <dspace:hasBitstream rdf:resource=""/> <dcterms:title>An Adaptive Multi Objective Selection Strategy for Active Learning</dcterms:title> <dcterms:isPartOf rdf:resource=""/> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dc:contributor>Cebron, Nicolas</dc:contributor> <dc:creator>Berthold, Michael R.</dc:creator> <dc:date rdf:datatype="">2011-03-24T16:09:49Z</dc:date> <dc:contributor>Berthold, Michael R.</dc:contributor> <dcterms:issued>2007</dcterms:issued> <dcterms:hasPart rdf:resource=""/> <dc:creator>Cebron, Nicolas</dc:creator> <dspace:isPartOfCollection rdf:resource=""/> <dcterms:rights rdf:resource=""/> <dc:format>application/pdf</dc:format> <dcterms:available rdf:datatype="">2011-03-24T16:09:49Z</dcterms:available> <dc:language>eng</dc:language> <dc:rights>terms-of-use</dc:rights> <bibo:uri rdf:resource=""/> <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 our previous approach and Active Learning with Support Vector Machines on several artificial and benchmark datasets.</dcterms:abstract> </rdf:Description> </rdf:RDF>

Dateiabrufe seit 01.10.2014 (Informationen über die Zugriffsstatistik)

preprint_235.pdf 35

Das Dokument erscheint in:

terms-of-use Solange nicht anders angezeigt, wird die Lizenz wie folgt beschrieben: terms-of-use

KOPS Suche


Mein Benutzerkonto