DataShiftExplorer : Visualizing and Comparing Change in Multidimensional Data for Supervised Learning

Schneider, Bruno; Keim, Daniel A.; El-Assady, Mennatallah

DataShiftExplorer : Visualizing and Comparing Change in Multidimensional Data for Supervised Learning

Dateien

Schneider_2-1sbj1x8la4jtz2.pdfGröße: 950.01 KBDownloads: 133

Datum

2020

Autor:innen

Schneider, Bruno

Keim, Daniel A.

El-Assady, Mennatallah

Publikationstyp

Beitrag zu einem Konferenzband

Publikationsstatus

Published

Erschienen in

KERREN, Andreas, ed. and others. VISIGRAPP 2020 : proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications : Valletta, Malta, February 27-29, 2020 ; Volume 3: IVAPP. Sétubal: SCITEPRESS, 2020, pp. 141-148. ISBN 9789897584022. Available under: doi: 10.5220/0008940801410148

Zusammenfassung

In supervised learning, to ensure the model's validity, it is essential to identify dataset shifts, i.e., when the data distribution changes from the one the model encountered at the time of training. To detect such changes, a comparative analysis of the multidimensional data distributions of the training data and new, unseen datasets is required. In this paper, we span the design space of visualizations for multidimensional comparative data analytics. Based on this design space, we present DataShiftExplorer, a technique tailored to identify and analyze the change in multidimensional data distributions. Throughout examples, we show how DataShiftExplorer facilitates the identification and analysis of data changes, supporting supervised learning.

Fachgebiet (DDC)

004 Informatik

Konferenz

15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2020, 27. Feb. 2020 - 29. Feb. 2020, Valletta, Malta

Zitieren

ISO 690

SCHNEIDER, Bruno, Daniel A. KEIM, Mennatallah EL-ASSADY, 2020. DataShiftExplorer : Visualizing and Comparing Change in Multidimensional Data for Supervised Learning. 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2020. Valletta, Malta, 27. Feb. 2020 - 29. Feb. 2020. In: KERREN, Andreas, ed. and others. VISIGRAPP 2020 : proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications : Valletta, Malta, February 27-29, 2020 ; Volume 3: IVAPP. Sétubal: SCITEPRESS, 2020, pp. 141-148. ISBN 9789897584022. Available under: doi: 10.5220/0008940801410148

BibTex

@inproceedings{Schneider2020-02-27DataS-52868,
  year={2020},
  doi={10.5220/0008940801410148},
  title={DataShiftExplorer : Visualizing and Comparing Change in Multidimensional Data for Supervised Learning},
  isbn={9789897584022},
  publisher={SCITEPRESS},
  address={Sétubal},
  booktitle={VISIGRAPP 2020 : proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications : Valletta, Malta, February 27-29, 2020 ; Volume 3: IVAPP},
  pages={141--148},
  editor={Kerren, Andreas},
  author={Schneider, Bruno and Keim, Daniel A. and El-Assady, Mennatallah}
}

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/123456789/52868">
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2021-02-17T09:45:10Z</dc:date>
    <dc:creator>El-Assady, Mennatallah</dc:creator>
    <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/52868"/>
    <dcterms:hasPart rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/52868/1/Schneider_2-1sbj1x8la4jtz2.pdf"/>
    <dcterms:abstract xml:lang="eng">In supervised learning, to ensure the model's validity, it is essential to identify dataset shifts, i.e., when the data distribution changes from the one the model encountered at the time of training. To detect such changes, a comparative analysis of the multidimensional data distributions of the training data and new, unseen datasets is required. In this paper, we span the design space of visualizations for multidimensional comparative data analytics. Based on this design space, we present DataShiftExplorer, a technique tailored to identify and analyze the change in multidimensional data distributions. Throughout examples, we show how DataShiftExplorer facilitates the identification and analysis of data changes, supporting supervised learning.</dcterms:abstract>
    <dc:contributor>Keim, Daniel A.</dc:contributor>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2021-02-17T09:45:10Z</dcterms:available>
    <dcterms:title>DataShiftExplorer : Visualizing and Comparing Change in Multidimensional Data for Supervised Learning</dcterms:title>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
    <dc:language>eng</dc:language>
    <dc:creator>Schneider, Bruno</dc:creator>
    <dspace:hasBitstream rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/52868/1/Schneider_2-1sbj1x8la4jtz2.pdf"/>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dcterms:rights rdf:resource="https://rightsstatements.org/page/InC/1.0/"/>
    <dc:contributor>El-Assady, Mennatallah</dc:contributor>
    <dc:contributor>Schneider, Bruno</dc:contributor>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <dc:creator>Keim, Daniel A.</dc:creator>
    <dc:rights>terms-of-use</dc:rights>
    <dcterms:issued>2020-02-27</dcterms:issued>
  </rdf:Description>
</rdf:RDF>

Universitätsbibliographie

Ja