Feature ranking through weights manipulations for artificial neural networks-based classifiers

Artificial Neural Networks (ANNs) are often viewed as black box. This limits the comprehensive understanding on how it deals with input neuron/data, as well as how it reached a particular decision. Input significance analysis (ISA) refers to the process of understanding these input neurons/data. An...

Full description

Bibliographic Details
Main Authors:	Hassan, Raini, Hassan, Wan Haslina, Alshaikhli, Imad Fakhri Taha, Ahmad, Salmiah, Alizadeh, Mojtaba
Other Authors:	Al-Dabass, David
Format:	Conference or Workshop Item
Language:	English English
Published:	The Institute of Electrical and Electronics Engineers 2014
Subjects:	T Technology (General)
Online Access:	http://irep.iium.edu.my/37854/ http://irep.iium.edu.my/37854/ http://irep.iium.edu.my/37854/ http://irep.iium.edu.my/37854/1/Feature_Ranking_Through_Weights_Manipulations_for_Artificial_Neural_Networks-.pdf http://irep.iium.edu.my/37854/4/37854.pdf

id	iium-37854
recordtype	eprints
spelling	iium-378542018-06-20T07:35:13Z http://irep.iium.edu.my/37854/ Feature ranking through weights manipulations for artificial neural networks-based classifiers Hassan, Raini Hassan, Wan Haslina Alshaikhli, Imad Fakhri Taha Ahmad, Salmiah Alizadeh, Mojtaba T Technology (General) Artificial Neural Networks (ANNs) are often viewed as black box. This limits the comprehensive understanding on how it deals with input neuron/data, as well as how it reached a particular decision. Input significance analysis (ISA) refers to the process of understanding these input neurons/data. And since this work is on classification problem, hence similarly, this process can also be called feature selection; where the goal is to have a classifier that can predict accurately and at the same time, its structure is as simple as possible. This work is particularly interested with ISA methods that manipulate weights, where separately, correlations are also applied. The goal is to create feature ranking list that performed the best in the selected classifiers. For validation methods, memory recall validation and K-Fold cross-validation methods are used. The results show one classifier that uses one of the ISA methods are performing well for both validation methods. The Institute of Electrical and Electronics Engineers Al-Dabass, David Sauli, Zaliman Zakaria, Zulkarnay 2014-01-27 Conference or Workshop Item PeerReviewed application/pdf en http://irep.iium.edu.my/37854/1/Feature_Ranking_Through_Weights_Manipulations_for_Artificial_Neural_Networks-.pdf application/pdf en http://irep.iium.edu.my/37854/4/37854.pdf Hassan, Raini and Hassan, Wan Haslina and Alshaikhli, Imad Fakhri Taha and Ahmad, Salmiah and Alizadeh, Mojtaba (2014) Feature ranking through weights manipulations for artificial neural networks-based classifiers. In: 2014 Fifth International Conference on Intelligent Systems, Modelling and Simulation (ISMS 2014), 27th-29th Jan. 2014, Langkawi, Kedah. http://ieeexplore.ieee.org/xpl/articleDetails.jsp?reload=true&arnumber=7280896 10.1109/ISMS.2014.31
repository_type	Digital Repository
institution_category	Local University
institution	International Islamic University Malaysia
building	IIUM Repository
collection	Online Access
language	English English
topic	T Technology (General)
spellingShingle	T Technology (General) Hassan, Raini Hassan, Wan Haslina Alshaikhli, Imad Fakhri Taha Ahmad, Salmiah Alizadeh, Mojtaba Feature ranking through weights manipulations for artificial neural networks-based classifiers
description	Artificial Neural Networks (ANNs) are often viewed as black box. This limits the comprehensive understanding on how it deals with input neuron/data, as well as how it reached a particular decision. Input significance analysis (ISA) refers to the process of understanding these input neurons/data. And since this work is on classification problem, hence similarly, this process can also be called feature selection; where the goal is to have a classifier that can predict accurately and at the same time, its structure is as simple as possible. This work is particularly interested with ISA methods that manipulate weights, where separately, correlations are also applied. The goal is to create feature ranking list that performed the best in the selected classifiers. For validation methods, memory recall validation and K-Fold cross-validation methods are used. The results show one classifier that uses one of the ISA methods are performing well for both validation methods.
author2	Al-Dabass, David
author_facet	Al-Dabass, David Hassan, Raini Hassan, Wan Haslina Alshaikhli, Imad Fakhri Taha Ahmad, Salmiah Alizadeh, Mojtaba
format	Conference or Workshop Item
author	Hassan, Raini Hassan, Wan Haslina Alshaikhli, Imad Fakhri Taha Ahmad, Salmiah Alizadeh, Mojtaba
author_sort	Hassan, Raini
title	Feature ranking through weights manipulations for artificial neural networks-based classifiers
title_short	Feature ranking through weights manipulations for artificial neural networks-based classifiers
title_full	Feature ranking through weights manipulations for artificial neural networks-based classifiers
title_fullStr	Feature ranking through weights manipulations for artificial neural networks-based classifiers
title_full_unstemmed	Feature ranking through weights manipulations for artificial neural networks-based classifiers
title_sort	feature ranking through weights manipulations for artificial neural networks-based classifiers
publisher	The Institute of Electrical and Electronics Engineers
publishDate	2014
url	http://irep.iium.edu.my/37854/ http://irep.iium.edu.my/37854/ http://irep.iium.edu.my/37854/ http://irep.iium.edu.my/37854/1/Feature_Ranking_Through_Weights_Manipulations_for_Artificial_Neural_Networks-.pdf http://irep.iium.edu.my/37854/4/37854.pdf
first_indexed	2023-09-18T20:54:16Z
last_indexed	2023-09-18T20:54:16Z
_version_	1777410199376625664

Feature ranking through weights manipulations for artificial neural networks-based classifiers

Similar Items