• Biometrical Letters

    Biometrical Letters

    Volume (52) Number 2 pp. 55-74

    Jolanta Grala-Michalak 1

    1Faculty of Mathematics and Computer Science, Adam Mickiewicz University, Umultowska 87, 61-614Poznań, Poland

    Directional representation of data in Linear Discriminant Analysis

    Abstract

    Sometimes feature representations of measured individuals are better described by spherical coordinates than Cartesian ones. The author proposes to introduce a preprocessing step in LDA based on the arctangent transformation of spherical coordinates. This nonlinear transformation does not change the dimension of the data, but in combination with LDA it leads to a dimension reduction if the raw data are not linearly separated. The method is presented using various examples of real and artificial data.

    Keywords: LDA, pattern recognition, spherical coordinates, dimension reduction, PCA, directional statistics
    https://content.sciendo.com/downloadpdf/journals/bile/52/2/article-p55.pdf Full text available in Adobe Acrobat format

    DOI: 10.1515/bile-2015-0006

    For citation:

    MLA Grala-Michalak, Jolanta. "Directional representation of data in Linear Discriminant Analysis." Biometrical Letters 52.2 (2015): 55-74. DOI: 10.1515/bile-2015-0006
    APA Grala-Michalak, J. (2015). Directional representation of data in Linear Discriminant Analysis. Biometrical Letters 52(2), 55-74 DOI: 10.1515/bile-2015-0006
    ISO 690 GRALA-MICHALAK, Jolanta. Directional representation of data in Linear Discriminant Analysis. Biometrical Letters, 2015, 52.2: 55-74. DOI: 10.1515/bile-2015-0006