Automated Feature Identification and Classification Using Automated Feature Weighted Self Organizing Map (FWSOM)

Andrew Starkey, Aliyu Usman Ahmad, Hassan Hamdoun

Research output: Contribution to journalArticle

1 Citation (Scopus)
9 Downloads (Pure)

Abstract

This paper investigates the application of a novel method for classification called Feature Weighted Self Organizing Map (FWSOM) that analyses the topology information of a converged standard Self Organizing Map (SOM) to automatically guide the selection of important inputs during training for improved classification of data with redundant inputs, examined against two traditional approaches namely neural networks and Support Vector Machines (SVM) for the classification of EEG data as presented in previous work. In particular, the novel method looks to identify the features that are important for classification automatically, and in this way the important features can be used to improve the diagnostic ability of any of the above methods. The paper presents the results and shows how the automated identification of the important features successfully identified the important features in the dataset and how this results in an improvement of the classification results for all methods apart from linear discriminatory methods which cannot separate the underlying nonlinear relationship in the data. The FWSOM in addition to achieving higher classification accuracy has given insights into what features are important in the classification of each class (left and right-hand movements), and these are corroborated by already published work in this area.

Original languageEnglish
Article number012006
Pages (from-to)1-7
Number of pages7
JournalIOP Conference Series: Materials Science and Engineering
Volume261
Issue number1
DOIs
Publication statusPublished - 6 Nov 2017
Event2017 International Conference on Artificial Intelligence Applications and Technologies (AIAAT 2017) - Hawaii, United States
Duration: 30 Aug 20172 Sep 2017

Fingerprint

Self organizing maps
Electroencephalography
Support vector machines
Topology
Neural networks

ASJC Scopus subject areas

  • Materials Science(all)
  • Engineering(all)

Cite this

Automated Feature Identification and Classification Using Automated Feature Weighted Self Organizing Map (FWSOM). / Starkey, Andrew; Ahmad, Aliyu Usman; Hamdoun, Hassan.

In: IOP Conference Series: Materials Science and Engineering, Vol. 261, No. 1, 012006, 06.11.2017, p. 1-7.

Research output: Contribution to journalArticle

@article{464828ebd417492e9405817ba8e9645e,
title = "Automated Feature Identification and Classification Using Automated Feature Weighted Self Organizing Map (FWSOM)",
abstract = "This paper investigates the application of a novel method for classification called Feature Weighted Self Organizing Map (FWSOM) that analyses the topology information of a converged standard Self Organizing Map (SOM) to automatically guide the selection of important inputs during training for improved classification of data with redundant inputs, examined against two traditional approaches namely neural networks and Support Vector Machines (SVM) for the classification of EEG data as presented in previous work. In particular, the novel method looks to identify the features that are important for classification automatically, and in this way the important features can be used to improve the diagnostic ability of any of the above methods. The paper presents the results and shows how the automated identification of the important features successfully identified the important features in the dataset and how this results in an improvement of the classification results for all methods apart from linear discriminatory methods which cannot separate the underlying nonlinear relationship in the data. The FWSOM in addition to achieving higher classification accuracy has given insights into what features are important in the classification of each class (left and right-hand movements), and these are corroborated by already published work in this area.",
author = "Andrew Starkey and Ahmad, {Aliyu Usman} and Hassan Hamdoun",
year = "2017",
month = "11",
day = "6",
doi = "10.1088/1757-899X/261/1/012006",
language = "English",
volume = "261",
pages = "1--7",
journal = "IOP Conference Series. Materials Science and Engineering",
issn = "1757-8981",
publisher = "IOP Publishing Ltd.",
number = "1",

}

TY - JOUR

T1 - Automated Feature Identification and Classification Using Automated Feature Weighted Self Organizing Map (FWSOM)

AU - Starkey, Andrew

AU - Ahmad, Aliyu Usman

AU - Hamdoun, Hassan

PY - 2017/11/6

Y1 - 2017/11/6

N2 - This paper investigates the application of a novel method for classification called Feature Weighted Self Organizing Map (FWSOM) that analyses the topology information of a converged standard Self Organizing Map (SOM) to automatically guide the selection of important inputs during training for improved classification of data with redundant inputs, examined against two traditional approaches namely neural networks and Support Vector Machines (SVM) for the classification of EEG data as presented in previous work. In particular, the novel method looks to identify the features that are important for classification automatically, and in this way the important features can be used to improve the diagnostic ability of any of the above methods. The paper presents the results and shows how the automated identification of the important features successfully identified the important features in the dataset and how this results in an improvement of the classification results for all methods apart from linear discriminatory methods which cannot separate the underlying nonlinear relationship in the data. The FWSOM in addition to achieving higher classification accuracy has given insights into what features are important in the classification of each class (left and right-hand movements), and these are corroborated by already published work in this area.

AB - This paper investigates the application of a novel method for classification called Feature Weighted Self Organizing Map (FWSOM) that analyses the topology information of a converged standard Self Organizing Map (SOM) to automatically guide the selection of important inputs during training for improved classification of data with redundant inputs, examined against two traditional approaches namely neural networks and Support Vector Machines (SVM) for the classification of EEG data as presented in previous work. In particular, the novel method looks to identify the features that are important for classification automatically, and in this way the important features can be used to improve the diagnostic ability of any of the above methods. The paper presents the results and shows how the automated identification of the important features successfully identified the important features in the dataset and how this results in an improvement of the classification results for all methods apart from linear discriminatory methods which cannot separate the underlying nonlinear relationship in the data. The FWSOM in addition to achieving higher classification accuracy has given insights into what features are important in the classification of each class (left and right-hand movements), and these are corroborated by already published work in this area.

UR - http://www.scopus.com/inward/record.url?scp=85035125017&partnerID=8YFLogxK

U2 - 10.1088/1757-899X/261/1/012006

DO - 10.1088/1757-899X/261/1/012006

M3 - Article

VL - 261

SP - 1

EP - 7

JO - IOP Conference Series. Materials Science and Engineering

JF - IOP Conference Series. Materials Science and Engineering

SN - 1757-8981

IS - 1

M1 - 012006

ER -