Control point selection for dimensionality reduction by radial basis function

Control point selection for dimensionality reduction by radial basis function

<p>This research deals with dimensionality reduction technique which is based on radial basis function (RBF) theory. The technique uses RBF for mapping multidimensional data points into a low-dimensional space by interpolating the previously calculated position of so-called control points. Thi...

Full description

Bibliographic Details
Main Authors: Kotryna Paulauskienė, Olga Kurasova
Format: Article
Language:English
Published: Klaipėda University 2016-02-01
Series:Computational Science and Techniques
Online Access:http://journals.ku.lt/index.php/CST/article/view/1095
id doaj-3e6cbe44d83c43fc88435207c0105293
record_format Article
spelling doaj-3e6cbe44d83c43fc88435207c01052932021-07-02T16:19:55ZengKlaipėda UniversityComputational Science and Techniques2029-99662016-02-014148749910.15181/csat.v4i1.10951105Control point selection for dimensionality reduction by radial basis functionKotryna Paulauskienė0Olga Kurasova1Vilnius University,Institute of Mathematics and InformaticsVilnius University, Institute of Mathematics and Informatics<p>This research deals with dimensionality reduction technique which is based on radial basis function (RBF) theory. The technique uses RBF for mapping multidimensional data points into a low-dimensional space by interpolating the previously calculated position of so-called control points. This paper analyses various ways of selection of control points (<em>regularized</em> <em>orthogonal least squares</em> method, <em>random</em> and <em>stratified</em> selections). The experiments have been carried out with 8 real and artificial data sets. Positions of the control points in a low-dimensional space are found by principal component analysis. We demonstrate that <em>random</em> and <em>stratified</em> selections of control points are efficient and acceptable in terms of balance between projection error (<em>stress</em>) and time-consumption.</p><p>DOI: 10.15181/csat.v4i1.1095</p>http://journals.ku.lt/index.php/CST/article/view/1095
collection DOAJ
language English
format Article
sources DOAJ
author Kotryna Paulauskienė
Olga Kurasova
spellingShingle Kotryna Paulauskienė
Olga Kurasova
Control point selection for dimensionality reduction by radial basis function
Computational Science and Techniques
author_facet Kotryna Paulauskienė
Olga Kurasova
author_sort Kotryna Paulauskienė
title Control point selection for dimensionality reduction by radial basis function
title_short Control point selection for dimensionality reduction by radial basis function
title_full Control point selection for dimensionality reduction by radial basis function
title_fullStr Control point selection for dimensionality reduction by radial basis function
title_full_unstemmed Control point selection for dimensionality reduction by radial basis function
title_sort control point selection for dimensionality reduction by radial basis function
publisher Klaipėda University
series Computational Science and Techniques
issn 2029-9966
publishDate 2016-02-01
description <p>This research deals with dimensionality reduction technique which is based on radial basis function (RBF) theory. The technique uses RBF for mapping multidimensional data points into a low-dimensional space by interpolating the previously calculated position of so-called control points. This paper analyses various ways of selection of control points (<em>regularized</em> <em>orthogonal least squares</em> method, <em>random</em> and <em>stratified</em> selections). The experiments have been carried out with 8 real and artificial data sets. Positions of the control points in a low-dimensional space are found by principal component analysis. We demonstrate that <em>random</em> and <em>stratified</em> selections of control points are efficient and acceptable in terms of balance between projection error (<em>stress</em>) and time-consumption.</p><p>DOI: 10.15181/csat.v4i1.1095</p>
url http://journals.ku.lt/index.php/CST/article/view/1095
work_keys_str_mv AT kotrynapaulauskiene controlpointselectionfordimensionalityreductionbyradialbasisfunction
AT olgakurasova controlpointselectionfordimensionalityreductionbyradialbasisfunction
_version_ 1721326786374008832

Similar Items