This article provides a high-level retrospective of type-2 fuzzy sets and fuzzy logic systems. It explains how type-2 fuzzy sets can be used to model membership function uncertainties, and how by doing this smoother performance can be obtained than by using type-1 fuzzy sets. It also summarizes the notation that should be used for type-2 fuzzy sets, describes four important mathematical representations for these fuzzy sets, explains the differences between type-1 and type-2 fuzzy logic systems and which of the four representations is most useful when designing an optimal type-2 fuzzy logic system, provides a very useful strategy for optimal designs of fuzzy logic systems -- one that guarantees performance improvement as one goes from a type-1 fuzzy logic system to a type-2 fuzzy logic system design -- , and describes four methods for simplifying the designs of type-2 fuzzy logic systems. Finally, it explains why type-2 fuzzy sets can capture two kinds of linguistic uncertainties simultaneously (the uncertainty of an individual and the uncertainties of a group about a word), whereas type-1 fuzzy sets cannot, and that such type-2 fuzzy set word models are what should be used to implement Zadeh's Computing With Words paradigm.
%0 Journal Article
%1 Mendel15insk
%A Mendel, Jerry M.
%D 2015
%J Informatik-Spektrum
%K 01614 springer paper ai knowledge processing fuzzy logic theory
%N 6
%P 523--532
%R 10.1007/s00287-015-0927-4
%T Type-2 Fuzzy Sets and Systems: a Retrospective
%V 38
%X This article provides a high-level retrospective of type-2 fuzzy sets and fuzzy logic systems. It explains how type-2 fuzzy sets can be used to model membership function uncertainties, and how by doing this smoother performance can be obtained than by using type-1 fuzzy sets. It also summarizes the notation that should be used for type-2 fuzzy sets, describes four important mathematical representations for these fuzzy sets, explains the differences between type-1 and type-2 fuzzy logic systems and which of the four representations is most useful when designing an optimal type-2 fuzzy logic system, provides a very useful strategy for optimal designs of fuzzy logic systems -- one that guarantees performance improvement as one goes from a type-1 fuzzy logic system to a type-2 fuzzy logic system design -- , and describes four methods for simplifying the designs of type-2 fuzzy logic systems. Finally, it explains why type-2 fuzzy sets can capture two kinds of linguistic uncertainties simultaneously (the uncertainty of an individual and the uncertainties of a group about a word), whereas type-1 fuzzy sets cannot, and that such type-2 fuzzy set word models are what should be used to implement Zadeh's Computing With Words paradigm.
@article{Mendel15insk,
abstract = {This article provides a high-level retrospective of type-2 fuzzy sets and fuzzy logic systems. It explains how type-2 fuzzy sets can be used to model membership function uncertainties, and how by doing this smoother performance can be obtained than by using type-1 fuzzy sets. It also summarizes the notation that should be used for type-2 fuzzy sets, describes four important mathematical representations for these fuzzy sets, explains the differences between type-1 and type-2 fuzzy logic systems and which of the four representations is most useful when designing an optimal type-2 fuzzy logic system, provides a very useful strategy for optimal designs of fuzzy logic systems -- one that guarantees performance improvement as one goes from a type-1 fuzzy logic system to a type-2 fuzzy logic system design -- , and describes four methods for simplifying the designs of type-2 fuzzy logic systems. Finally, it explains why type-2 fuzzy sets can capture two kinds of linguistic uncertainties simultaneously (the uncertainty of an individual and the uncertainties of a group about a word), whereas type-1 fuzzy sets cannot, and that such type-2 fuzzy set word models are what should be used to implement Zadeh's Computing With Words paradigm.},
added-at = {2016-05-22T09:53:02.000+0200},
author = {Mendel, Jerry M.},
biburl = {https://www.bibsonomy.org/bibtex/2276d4d84456c0a1b40e43bf9598dd58d/flint63},
doi = {10.1007/s00287-015-0927-4},
file = {SpringerLink:2015/Mendel15insk.pdf:PDF},
groups = {public},
interhash = {31e2d501bc1fcbf882470ab31cade528},
intrahash = {276d4d84456c0a1b40e43bf9598dd58d},
issn = {0170-6012},
journal = {Informatik-Spektrum},
keywords = {01614 springer paper ai knowledge processing fuzzy logic theory},
month = {#dec#},
number = 6,
pages = {523--532},
timestamp = {2018-04-16T12:08:45.000+0200},
title = {Type-2 {Fuzzy} Sets and Systems: a Retrospective},
username = {flint63},
volume = 38,
year = 2015
}