UCL logo

UCL Discovery

UCL home » Library Services » Electronic resources » UCL Discovery

On the orthogonal distance to class subspaces for high-dimensional data classification

Zhu, R; Xue, J-H; (2017) On the orthogonal distance to class subspaces for high-dimensional data classification. Information Sciences , 417 pp. 262-273. 10.1016/j.ins.2017.07.019. Green open access

[img]
Preview
Text
RuiZhu-INS-2017-UCL.pdf - ["content_typename_Accepted version" not defined]

Download (328kB) | Preview

Abstract

The orthogonal distance from an instance to the subspace of a class is a key metric for pattern classification by the class subspace-based methods. There is a close relationship between the orthogonal distance and the residual standard deviation of a test instance from the class subspace. In this paper, we shall show that an established and widely-used relationship, between the residual standard deviation and the sum of squares of the residual PC scores, is not precise, and thus can lead to incorrect results, for the inference of high-dimensional data which nowadays are common in practice.

Type: Article
Title: On the orthogonal distance to class subspaces for high-dimensional data classification
Open access status: An open access version is available from UCL Discovery
DOI: 10.1016/j.ins.2017.07.019
Publisher version: https://doi.org/10.1016/j.ins.2017.07.019
Language: English
Additional information: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.
Keywords: Classification; High-dimensional data; Orthogonal distance; Principal component analysis (PCA); Soft independent modelling of class analogy (SIMCA)
UCL classification: UCL > Provost and Vice Provost Offices
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences > Dept of Statistical Science
URI: http://discovery.ucl.ac.uk/id/eprint/1565596
Downloads since deposit
55Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item