A Rule Extraction Algorithm That Scales Between Fidelity and Comprehensibility

Fidelity and comprehensibility are the common measures used in the evaluation of rules extracted from neural networks. However, these two measures are found to be inverse relations of one another. Since the needs of comprehensibility or fidelity may vary depending on the user or application, this pa...

Full description

Bibliographic Details
Main Authors: Muthu Anbananthen, Kalaiarasi Sonai, Chan, Fabian Huan Pheng*, Subramaniam, Subhacini, Eimad Eldin, Abdu Ali Abusham
Format: Article
Language:English
Published: Science Alert 2012
Subjects:
Online Access:http://eprints.intimal.edu.my/1003/
http://eprints.intimal.edu.my/1003/1/A%20rule%20extraction%20algorithm%20that%20scales%20between%20fidelity%20and%20comprehensibility.pdf
_version_ 1848766618966949888
author Muthu Anbananthen, Kalaiarasi Sonai
Chan, Fabian Huan Pheng*
Subramaniam, Subhacini
Eimad Eldin, Abdu Ali Abusham
author_facet Muthu Anbananthen, Kalaiarasi Sonai
Chan, Fabian Huan Pheng*
Subramaniam, Subhacini
Eimad Eldin, Abdu Ali Abusham
author_sort Muthu Anbananthen, Kalaiarasi Sonai
building INTI Institutional Repository
collection Online Access
description Fidelity and comprehensibility are the common measures used in the evaluation of rules extracted from neural networks. However, these two measures are found to be inverse relations of one another. Since the needs of comprehensibility or fidelity may vary depending on the user or application, this paper presented a significance based rule extraction algorithm that allows a user set parameter to scale between the desired degree of fidelity and comprehensibility of the rules extracted. A detailed explanation and example application of this algorithm is presented as well as experimental results on several neural network ensembles.
first_indexed 2025-11-14T11:54:01Z
format Article
id intimal-1003
institution INTI International University
institution_category Local University
language English
last_indexed 2025-11-14T11:54:01Z
publishDate 2012
publisher Science Alert
recordtype eprints
repository_type Digital Repository
spelling intimal-10032017-12-04T02:58:59Z http://eprints.intimal.edu.my/1003/ A Rule Extraction Algorithm That Scales Between Fidelity and Comprehensibility Muthu Anbananthen, Kalaiarasi Sonai Chan, Fabian Huan Pheng* Subramaniam, Subhacini Eimad Eldin, Abdu Ali Abusham QA76 Computer software Fidelity and comprehensibility are the common measures used in the evaluation of rules extracted from neural networks. However, these two measures are found to be inverse relations of one another. Since the needs of comprehensibility or fidelity may vary depending on the user or application, this paper presented a significance based rule extraction algorithm that allows a user set parameter to scale between the desired degree of fidelity and comprehensibility of the rules extracted. A detailed explanation and example application of this algorithm is presented as well as experimental results on several neural network ensembles. Science Alert 2012 Article PeerReviewed text en http://eprints.intimal.edu.my/1003/1/A%20rule%20extraction%20algorithm%20that%20scales%20between%20fidelity%20and%20comprehensibility.pdf Muthu Anbananthen, Kalaiarasi Sonai and Chan, Fabian Huan Pheng* and Subramaniam, Subhacini and Eimad Eldin, Abdu Ali Abusham (2012) A Rule Extraction Algorithm That Scales Between Fidelity and Comprehensibility. Asian Journal of Scientific Research, 5 (3). pp. 121-132. ISSN 2077-2076 http://scialert.net/abstract/?doi=ajsr.2012.121.132 10.3923/ajsr.2012.121.132
spellingShingle QA76 Computer software
Muthu Anbananthen, Kalaiarasi Sonai
Chan, Fabian Huan Pheng*
Subramaniam, Subhacini
Eimad Eldin, Abdu Ali Abusham
A Rule Extraction Algorithm That Scales Between Fidelity and Comprehensibility
title A Rule Extraction Algorithm That Scales Between Fidelity and Comprehensibility
title_full A Rule Extraction Algorithm That Scales Between Fidelity and Comprehensibility
title_fullStr A Rule Extraction Algorithm That Scales Between Fidelity and Comprehensibility
title_full_unstemmed A Rule Extraction Algorithm That Scales Between Fidelity and Comprehensibility
title_short A Rule Extraction Algorithm That Scales Between Fidelity and Comprehensibility
title_sort rule extraction algorithm that scales between fidelity and comprehensibility
topic QA76 Computer software
url http://eprints.intimal.edu.my/1003/
http://eprints.intimal.edu.my/1003/
http://eprints.intimal.edu.my/1003/
http://eprints.intimal.edu.my/1003/1/A%20rule%20extraction%20algorithm%20that%20scales%20between%20fidelity%20and%20comprehensibility.pdf