Please use this identifier to cite or link to this item:
https://hdl.handle.net/20.500.11851/7771
Title: | Upper and lower values for the level of fuzziness in FCM | Authors: | Özkan, İbrahim Türkşen, İsmail Burhan |
Keywords: | [No Keywords] | Publisher: | Elsevier Science Inc | Abstract: | The level of fuzziness is a parameter in fuzzy system modeling which is a source of uncertainty. In order to explore the effect of this uncertainty, one needs to investigate and identify effective upper and lower boundaries of the level of fuzziness. For this purpose, Fuzzy c-means (FCM) clustering methodology is investigated to determine the effective upper and lower boundaries of the level of fuzziness in order to capture the uncertainty generated by this parameter. In this regard, we propose to expand the membership function around important information points of FCM. These important information points are, cluster centers and the mass center. At these points, it is known that, the level of fuzziness has no effect on the membership values. In this way, we identify the counter-intuitive behavior of membership function near these particular information points. It will be shown that the upper and lower values of the level of fuzziness can be identified. Hence the uncertainty generated by this parameter can be encapsulated. (C) 2007 Elsevier Inc. All rights reserved. | URI: | https://doi.org/10.1016/j.ins.2007.06.028 https://hdl.handle.net/20.500.11851/7771 |
ISSN: | 0020-0255 |
Appears in Collections: | Endüstri Mühendisliği Bölümü / Department of Industrial Engineering Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection |
Show full item record
CORE Recommender
SCOPUSTM
Citations
79
checked on Nov 2, 2024
WEB OF SCIENCETM
Citations
67
checked on Nov 2, 2024
Page view(s)
78
checked on Nov 4, 2024
Google ScholarTM
Check
Altmetric
Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.