Research output

Conference Paper (in Formal Publication) ()

Evolutionary Algorithms with Linkage Information for Feature Selection in Brain Computer Interfaces

Citation
Adair J, Brownlee A & Ochoa G (2016) Evolutionary Algorithms with Linkage Information for Feature Selection in Brain Computer Interfaces In: Angelov P, Gegov A, Jayne C, Shen Q (ed.) Advances in Computational Intelligence Systems: Contributions Presented at the 16th UK Workshop on Computational Intelligence, September 7–9, 2016, Lancaster, UK, London: Springer. UKCI 2016 - 16th UK Workshop on Computational Intelligence, 7.9.2016 - 9.9.2016, Lancaster, pp. 287-307.

Abstract
Abstract Brain Computer Interfaces are an essential technology for the advancement of prosthetic limbs, but current signal acquisition methods are hindered by a number of factors, not least, noise. In this context, Feature Selection is required to choose the important signal features and improve classifier accuracy. Evolutionary algorithms have proven to outperform filtering methods (in terms of accuracy) for Feature Selection. This paper applies a single-point heuristic search method, Iterated Local Search (ILS), and compares it to a genetic algorithm (GA) and a memetic algorithm (MA). It then further attempts to utilise Linkage between features to guide search operators in the algorithms stated. The GA was found to outperform ILS. Counter-intuitively, linkage-guided algorithms resulted in higher classification error rates than their unguided alternatives. Explanations for this are explored.

Keywords
evolutionary search; brain computer interfaces; Iterated Local Search; Genetic Algorithms; Feature Selection; Intelligent Operators; feature selection; memetic algorithms; linkage score; linkage detection algorithms; epistasis; eeg; prosthetics

StatusPublished
EditorAngelov P, Gegov A, Jayne C, Shen Q
AuthorsAdair Jason, Brownlee Alexander, Ochoa Gabriela
Title of seriesAdvances in Intelligent Systems and Computing
Number in series513
Publication date07/09/2016
Date of public distribution09/2016
Date accepted by journal01/08/2016
URLhttp://link.springer.com/…3-319-46562-3_19
PublisherSpringer
Place of publicationLondon
ISSN of series 2194-5357
ISBN 978-3-319-46561-6
eISBN 978-3-319-46562-3
LanguageEnglish
© University of Stirling FK9 4LA Scotland UK • Telephone +44 1786 473171 • Scottish Charity No SC011159
My Portal