University of Surrey

Test tubes in the lab Research in the ATI Dance Research

Optimising Ensemble of Two-Class classifiers using Spectral Analysis

Windeatt, Terry (2018) Optimising Ensemble of Two-Class classifiers using Spectral Analysis In: 24th International Conference on Pattern Recognition (ICPR) 2018, 20 - 24 August 2018, Beijing, China.

[img]
Preview
Text
PID5352217.pdf - Accepted version Manuscript

Download (1MB) | Preview

Abstract

An approach to approximating the decision boundary of an ensemble of two-class classifiers is proposed. Spectral coefficients are used to approximate the discrete probability density function of a Boolean Function. It is shown that the difference between first and third order coefficient approximation is a good indicator of optimal base classifier complexity. A theoretical analysis is supported by experimental results on a variety of Artificial and Real two-class problems.

Item Type: Conference or Workshop Item (Conference Paper)
Divisions : Faculty of Engineering and Physical Sciences > Electronic Engineering
Authors :
NameEmailORCID
Windeatt, Terryt.windeatt@surrey.ac.uk
Date : 2018
Copyright Disclaimer : © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Uncontrolled Keywords : Boolean functions, ensemble classifier, multilayer perceptrons, pattern analysis, spectral analysis, supervised learning
Depositing User : Melanie Hughes
Date Deposited : 02 Aug 2018 08:14
Last Modified : 24 Aug 2018 02:08
URI: http://epubs.surrey.ac.uk/id/eprint/848830

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year


Information about this web site

© The University of Surrey, Guildford, Surrey, GU2 7XH, United Kingdom.
+44 (0)1483 300800