Feature scaling in support vector data description

P Juszczak, DMJ Tax, RPW Duin

    Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientific


    When in a classification problem only samples of one class are easily accessible, this problem is called a one-class classification problem. Many standard classifiers, like backpropagation neural networks, fail on this data. Some other classifiers, like k-means clustering or nearest neighbor classifier can be applied after some minor changes. In this paper we focus on the support vector data description classifier, which is especially constructed for one-class classification. But this method appears to be sensitive to scaling of the individual features of the dataset. We show that it is possible to improve its performance by adequate scaling of the feature space. Some results will be shown on artificial dataset and handwritten digits dataset.
    Original languageUndefined/Unknown
    Title of host publicationProceedings ASCI 2002
    EditorsEF Deprettere, A Belloum, JWJ Heijnsdijk, F van der Stappen
    Place of PublicationDelft
    Number of pages8
    ISBN (Print)90-803086-6-8
    Publication statusPublished - 2002
    Event8th Annual Conf. of the Advanced School for Computing and Imaging, Lochem - Delft
    Duration: 19 Jun 200221 Jun 2002

    Publication series



    Conference8th Annual Conf. of the Advanced School for Computing and Imaging, Lochem

    Bibliographical note

    phpub 19


    • conference contrib. non-refer.
    • Geen BTA classificatie

    Cite this