No Cover Image

Journal article 869 views 247 downloads

Extracting Takagi-Sugeno Fuzzy Rules with Interpretable Submodels via Regularization of Linguistic Modifiers

Shang-Ming Zhou, J.Q. Gan, Shang-ming Zhou Orcid Logo

IEEE Transactions on Knowledge and Data Engineering, Volume: 21, Issue: 8, Pages: 1191 - 1204

Swansea University Author: Shang-ming Zhou Orcid Logo

Check full text

DOI (Published version): 10.1109/TKDE.2008.208

Abstract

In this paper, a method for constructing Takagi-Sugeno (TS) fuzzy system from data is proposed with the objective of preserving TS submodel comprehensibility, in which linguistic modifiers are suggested to characterize the fuzzy sets. A good property held by the proposed linguistic modifiers is that...

Full description

Published in: IEEE Transactions on Knowledge and Data Engineering
ISSN: 1041-4347
Published: IEEE Transactions on Knowledge and Data Engineering 2008
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa10026
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2013-07-23T12:02:39Z
last_indexed 2019-07-17T19:49:08Z
id cronfa10026
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2019-07-17T15:00:45.3052670</datestamp><bib-version>v2</bib-version><id>10026</id><entry>2012-03-21</entry><title>Extracting Takagi-Sugeno Fuzzy Rules with Interpretable Submodels via Regularization of Linguistic Modifiers</title><swanseaauthors><author><sid>118578a62021ba8ef61398da0a8750da</sid><ORCID>0000-0002-0719-9353</ORCID><firstname>Shang-ming</firstname><surname>Zhou</surname><name>Shang-ming Zhou</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2012-03-21</date><deptcode>BMS</deptcode><abstract>In this paper, a method for constructing Takagi-Sugeno (TS) fuzzy system from data is proposed with the objective of preserving TS submodel comprehensibility, in which linguistic modifiers are suggested to characterize the fuzzy sets. A good property held by the proposed linguistic modifiers is that they can broaden the cores of fuzzy sets while contracting the overlaps of adjoining membership functions (MFs) during identification of fuzzy systems from data. As a result, the TS submodels identified tend to dominate the system behaviors by automatically matching the global model (GM) in corresponding subareas, which leads to good TS model interpretability while producing distinguishable input space partitioning. However, the GM accuracy and model interpretability are two conflicting modeling objectives, improving interpretability of fuzzy models generally degrades the GM performance of fuzzy models, and vice versa. Hence, one challenging problem is how to construct a TS fuzzy model with not only good global performance but also good submodel interpretability. In order to achieve a good tradeoff between GM performance and submodel interpretability, a regularization learning algorithm is presented in which the GM objective function is combined with a local model objective function defined in terms of an extended index of fuzziness of identified MFs. Moreover, a parsimonious rule base is obtained by adopting a QR decomposition method to select the important fuzzy rules and reduce the redundant ones. Experimental studies have shown that the TS models identified by the suggested method possess good submodel interpretability and satisfactory GM performance with parsimonious rule bases.</abstract><type>Journal Article</type><journal>IEEE Transactions on Knowledge and Data Engineering</journal><volume>21</volume><journalNumber>8</journalNumber><paginationStart>1191</paginationStart><paginationEnd>1204</paginationEnd><publisher>IEEE Transactions on Knowledge and Data Engineering</publisher><issnPrint>1041-4347</issnPrint><keywords>Interpretability , Knowledge extraction , Takagi-Sugeno fuzzy models , comprehensibility , distinguishability , fuzziness. , knowledge extraction , linearization , local linear models , local models , regularization , submodels , transparency</keywords><publishedDay>10</publishedDay><publishedMonth>10</publishedMonth><publishedYear>2008</publishedYear><publishedDate>2008-10-10</publishedDate><doi>10.1109/TKDE.2008.208</doi><url/><notes/><college>COLLEGE NANME</college><department>Biomedical Sciences</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>BMS</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2019-07-17T15:00:45.3052670</lastEdited><Created>2012-03-21T16:17:09.0000000</Created><path><level id="1">Faculty of Medicine, Health and Life Sciences</level><level id="2">Swansea University Medical School - Medicine</level></path><authors><author><firstname/><surname>Shang-Ming Zhou</surname><order>1</order></author><author><firstname>J.Q.</firstname><surname>Gan</surname><order>2</order></author><author><firstname>Shang-ming</firstname><surname>Zhou</surname><orcid>0000-0002-0719-9353</orcid><order>3</order></author></authors><documents><document><filename>0010026-26042019163230.pdf</filename><originalFilename>TKDE-2008-03-0149-2OK.pdf</originalFilename><uploaded>2019-04-26T16:32:30.0670000</uploaded><type>Output</type><contentLength>1288163</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><embargoDate>2019-04-26T00:00:00.0000000</embargoDate><copyrightCorrect>true</copyrightCorrect><language>eng</language></document></documents><OutputDurs/></rfc1807>
spelling 2019-07-17T15:00:45.3052670 v2 10026 2012-03-21 Extracting Takagi-Sugeno Fuzzy Rules with Interpretable Submodels via Regularization of Linguistic Modifiers 118578a62021ba8ef61398da0a8750da 0000-0002-0719-9353 Shang-ming Zhou Shang-ming Zhou true false 2012-03-21 BMS In this paper, a method for constructing Takagi-Sugeno (TS) fuzzy system from data is proposed with the objective of preserving TS submodel comprehensibility, in which linguistic modifiers are suggested to characterize the fuzzy sets. A good property held by the proposed linguistic modifiers is that they can broaden the cores of fuzzy sets while contracting the overlaps of adjoining membership functions (MFs) during identification of fuzzy systems from data. As a result, the TS submodels identified tend to dominate the system behaviors by automatically matching the global model (GM) in corresponding subareas, which leads to good TS model interpretability while producing distinguishable input space partitioning. However, the GM accuracy and model interpretability are two conflicting modeling objectives, improving interpretability of fuzzy models generally degrades the GM performance of fuzzy models, and vice versa. Hence, one challenging problem is how to construct a TS fuzzy model with not only good global performance but also good submodel interpretability. In order to achieve a good tradeoff between GM performance and submodel interpretability, a regularization learning algorithm is presented in which the GM objective function is combined with a local model objective function defined in terms of an extended index of fuzziness of identified MFs. Moreover, a parsimonious rule base is obtained by adopting a QR decomposition method to select the important fuzzy rules and reduce the redundant ones. Experimental studies have shown that the TS models identified by the suggested method possess good submodel interpretability and satisfactory GM performance with parsimonious rule bases. Journal Article IEEE Transactions on Knowledge and Data Engineering 21 8 1191 1204 IEEE Transactions on Knowledge and Data Engineering 1041-4347 Interpretability , Knowledge extraction , Takagi-Sugeno fuzzy models , comprehensibility , distinguishability , fuzziness. , knowledge extraction , linearization , local linear models , local models , regularization , submodels , transparency 10 10 2008 2008-10-10 10.1109/TKDE.2008.208 COLLEGE NANME Biomedical Sciences COLLEGE CODE BMS Swansea University 2019-07-17T15:00:45.3052670 2012-03-21T16:17:09.0000000 Faculty of Medicine, Health and Life Sciences Swansea University Medical School - Medicine Shang-Ming Zhou 1 J.Q. Gan 2 Shang-ming Zhou 0000-0002-0719-9353 3 0010026-26042019163230.pdf TKDE-2008-03-0149-2OK.pdf 2019-04-26T16:32:30.0670000 Output 1288163 application/pdf Accepted Manuscript true 2019-04-26T00:00:00.0000000 true eng
title Extracting Takagi-Sugeno Fuzzy Rules with Interpretable Submodels via Regularization of Linguistic Modifiers
spellingShingle Extracting Takagi-Sugeno Fuzzy Rules with Interpretable Submodels via Regularization of Linguistic Modifiers
Shang-ming Zhou
title_short Extracting Takagi-Sugeno Fuzzy Rules with Interpretable Submodels via Regularization of Linguistic Modifiers
title_full Extracting Takagi-Sugeno Fuzzy Rules with Interpretable Submodels via Regularization of Linguistic Modifiers
title_fullStr Extracting Takagi-Sugeno Fuzzy Rules with Interpretable Submodels via Regularization of Linguistic Modifiers
title_full_unstemmed Extracting Takagi-Sugeno Fuzzy Rules with Interpretable Submodels via Regularization of Linguistic Modifiers
title_sort Extracting Takagi-Sugeno Fuzzy Rules with Interpretable Submodels via Regularization of Linguistic Modifiers
author_id_str_mv 118578a62021ba8ef61398da0a8750da
author_id_fullname_str_mv 118578a62021ba8ef61398da0a8750da_***_Shang-ming Zhou
author Shang-ming Zhou
author2 Shang-Ming Zhou
J.Q. Gan
Shang-ming Zhou
format Journal article
container_title IEEE Transactions on Knowledge and Data Engineering
container_volume 21
container_issue 8
container_start_page 1191
publishDate 2008
institution Swansea University
issn 1041-4347
doi_str_mv 10.1109/TKDE.2008.208
publisher IEEE Transactions on Knowledge and Data Engineering
college_str Faculty of Medicine, Health and Life Sciences
hierarchytype
hierarchy_top_id facultyofmedicinehealthandlifesciences
hierarchy_top_title Faculty of Medicine, Health and Life Sciences
hierarchy_parent_id facultyofmedicinehealthandlifesciences
hierarchy_parent_title Faculty of Medicine, Health and Life Sciences
department_str Swansea University Medical School - Medicine{{{_:::_}}}Faculty of Medicine, Health and Life Sciences{{{_:::_}}}Swansea University Medical School - Medicine
document_store_str 1
active_str 0
description In this paper, a method for constructing Takagi-Sugeno (TS) fuzzy system from data is proposed with the objective of preserving TS submodel comprehensibility, in which linguistic modifiers are suggested to characterize the fuzzy sets. A good property held by the proposed linguistic modifiers is that they can broaden the cores of fuzzy sets while contracting the overlaps of adjoining membership functions (MFs) during identification of fuzzy systems from data. As a result, the TS submodels identified tend to dominate the system behaviors by automatically matching the global model (GM) in corresponding subareas, which leads to good TS model interpretability while producing distinguishable input space partitioning. However, the GM accuracy and model interpretability are two conflicting modeling objectives, improving interpretability of fuzzy models generally degrades the GM performance of fuzzy models, and vice versa. Hence, one challenging problem is how to construct a TS fuzzy model with not only good global performance but also good submodel interpretability. In order to achieve a good tradeoff between GM performance and submodel interpretability, a regularization learning algorithm is presented in which the GM objective function is combined with a local model objective function defined in terms of an extended index of fuzziness of identified MFs. Moreover, a parsimonious rule base is obtained by adopting a QR decomposition method to select the important fuzzy rules and reduce the redundant ones. Experimental studies have shown that the TS models identified by the suggested method possess good submodel interpretability and satisfactory GM performance with parsimonious rule bases.
published_date 2008-10-10T03:10:38Z
_version_ 1763749961562849280
score 11.037603

Similar Items