Comparison of methods for automated feature selection using a Self-Organising Map

Aliyu Usman Ahmad, Andrew Starkey

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The effective modelling of high-dimensional data with hundreds to thousands of features remains a challenging task in the field of machine learning. One of the key challenges is the implementation of effective methods for selecting a set of relevant features, which are buried in high-dimensional data along with irrelevant noisy features by choosing a subset of the complete set of input features that predicts the output with higher accuracy comparable to the performance of the complete input set. Kohonen’s Self Organising Neural Network MAP has been utilized in various ways for this task. In this work, a review of the appropriate application of multiple methods for this task is carried out. The feature selection approach based on analysis of the Self Organising network result after training is presented with comparison of performance of two methods.
Original languageEnglish
Title of host publicationEngineering Applications of Neural Networks
Subtitle of host publication17th International Conference, EANN 2016, Aberdeen, UK, September 2-5, 2016, Proceedings
EditorsChrisina Jayne, Lazaros Iliadis
PublisherSpringer-Verlag
Pages134-146
Number of pages13
VolumeCCIS 269
ISBN (Electronic)978-3-319-44188-7
ISBN (Print)978-3-319-44187-0
DOIs
Publication statusPublished - 2016

Publication series

NameCommunications in Computer and Information Science
PublisherSpringer
Volume629
ISSN (Print)1865-0929

Fingerprint

Self organizing maps
Learning systems
Feature extraction
Neural networks

Keywords

  • Clustering
  • Self-organising neural network MAP
  • Feature selection
  • Engineering optimisation

Cite this

Ahmad, A. U., & Starkey, A. (2016). Comparison of methods for automated feature selection using a Self-Organising Map. In C. Jayne, & L. Iliadis (Eds.), Engineering Applications of Neural Networks: 17th International Conference, EANN 2016, Aberdeen, UK, September 2-5, 2016, Proceedings (Vol. CCIS 269, pp. 134-146). (Communications in Computer and Information Science; Vol. 629). Springer-Verlag. https://doi.org/10.1007/978-3-319-44188-7_10

Comparison of methods for automated feature selection using a Self-Organising Map. / Ahmad, Aliyu Usman; Starkey, Andrew.

Engineering Applications of Neural Networks: 17th International Conference, EANN 2016, Aberdeen, UK, September 2-5, 2016, Proceedings. ed. / Chrisina Jayne; Lazaros Iliadis. Vol. CCIS 269 Springer-Verlag, 2016. p. 134-146 (Communications in Computer and Information Science; Vol. 629).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ahmad, AU & Starkey, A 2016, Comparison of methods for automated feature selection using a Self-Organising Map. in C Jayne & L Iliadis (eds), Engineering Applications of Neural Networks: 17th International Conference, EANN 2016, Aberdeen, UK, September 2-5, 2016, Proceedings. vol. CCIS 269, Communications in Computer and Information Science, vol. 629, Springer-Verlag, pp. 134-146. https://doi.org/10.1007/978-3-319-44188-7_10
Ahmad AU, Starkey A. Comparison of methods for automated feature selection using a Self-Organising Map. In Jayne C, Iliadis L, editors, Engineering Applications of Neural Networks: 17th International Conference, EANN 2016, Aberdeen, UK, September 2-5, 2016, Proceedings. Vol. CCIS 269. Springer-Verlag. 2016. p. 134-146. (Communications in Computer and Information Science). https://doi.org/10.1007/978-3-319-44188-7_10
Ahmad, Aliyu Usman ; Starkey, Andrew. / Comparison of methods for automated feature selection using a Self-Organising Map. Engineering Applications of Neural Networks: 17th International Conference, EANN 2016, Aberdeen, UK, September 2-5, 2016, Proceedings. editor / Chrisina Jayne ; Lazaros Iliadis. Vol. CCIS 269 Springer-Verlag, 2016. pp. 134-146 (Communications in Computer and Information Science).
@inproceedings{c8ada8e800e64c66a7cbde9285c25054,
title = "Comparison of methods for automated feature selection using a Self-Organising Map",
abstract = "The effective modelling of high-dimensional data with hundreds to thousands of features remains a challenging task in the field of machine learning. One of the key challenges is the implementation of effective methods for selecting a set of relevant features, which are buried in high-dimensional data along with irrelevant noisy features by choosing a subset of the complete set of input features that predicts the output with higher accuracy comparable to the performance of the complete input set. Kohonen’s Self Organising Neural Network MAP has been utilized in various ways for this task. In this work, a review of the appropriate application of multiple methods for this task is carried out. The feature selection approach based on analysis of the Self Organising network result after training is presented with comparison of performance of two methods.",
keywords = "Clustering, Self-organising neural network MAP, Feature selection, Engineering optimisation",
author = "Ahmad, {Aliyu Usman} and Andrew Starkey",
year = "2016",
doi = "10.1007/978-3-319-44188-7_10",
language = "English",
isbn = "978-3-319-44187-0",
volume = "CCIS 269",
series = "Communications in Computer and Information Science",
publisher = "Springer-Verlag",
pages = "134--146",
editor = "Chrisina Jayne and Lazaros Iliadis",
booktitle = "Engineering Applications of Neural Networks",

}

TY - GEN

T1 - Comparison of methods for automated feature selection using a Self-Organising Map

AU - Ahmad, Aliyu Usman

AU - Starkey, Andrew

PY - 2016

Y1 - 2016

N2 - The effective modelling of high-dimensional data with hundreds to thousands of features remains a challenging task in the field of machine learning. One of the key challenges is the implementation of effective methods for selecting a set of relevant features, which are buried in high-dimensional data along with irrelevant noisy features by choosing a subset of the complete set of input features that predicts the output with higher accuracy comparable to the performance of the complete input set. Kohonen’s Self Organising Neural Network MAP has been utilized in various ways for this task. In this work, a review of the appropriate application of multiple methods for this task is carried out. The feature selection approach based on analysis of the Self Organising network result after training is presented with comparison of performance of two methods.

AB - The effective modelling of high-dimensional data with hundreds to thousands of features remains a challenging task in the field of machine learning. One of the key challenges is the implementation of effective methods for selecting a set of relevant features, which are buried in high-dimensional data along with irrelevant noisy features by choosing a subset of the complete set of input features that predicts the output with higher accuracy comparable to the performance of the complete input set. Kohonen’s Self Organising Neural Network MAP has been utilized in various ways for this task. In this work, a review of the appropriate application of multiple methods for this task is carried out. The feature selection approach based on analysis of the Self Organising network result after training is presented with comparison of performance of two methods.

KW - Clustering

KW - Self-organising neural network MAP

KW - Feature selection

KW - Engineering optimisation

U2 - 10.1007/978-3-319-44188-7_10

DO - 10.1007/978-3-319-44188-7_10

M3 - Conference contribution

SN - 978-3-319-44187-0

VL - CCIS 269

T3 - Communications in Computer and Information Science

SP - 134

EP - 146

BT - Engineering Applications of Neural Networks

A2 - Jayne, Chrisina

A2 - Iliadis, Lazaros

PB - Springer-Verlag

ER -