Interval-valued fuzzy associative memories based on representable conjunctions with applications in prediction

Author

Sussner, Peter ; Schuster, Thomas

Author_Institution

Dept. of Appl. Math., Univ. of Campinas, Campinas, Brazil

fYear

2013

fDate

24-28 June 2013

Firstpage

344

Lastpage

349

Abstract

The mathematical foundations of fuzzy morphological associative memories (FMAMs) can be found in mathematical morphology on complete lattices that include the class of fuzzy sets over a certain universe as a special case. The advent of type-2 fuzzy systems suggests the development of type-2 FMAMs and in particular interval-valued (type-2) FMAMs since the classes of general type-2 fuzzy sets and interval-valued fuzzy sets over a certain universe form complete lattices. For the sake of mathematical and computational simplicity, we employ representable conjunctions of interval-valued fuzzy set theory to define interval-valued FMAMs (IV-FMAMs) in this paper. Finally, we apply the resulting IV-FMAM models to a problem of monthly streamflow prediction of a large Brazilian hydro-electric plant.

Keywords

content-addressable storage; fuzzy neural nets; fuzzy set theory; lattice theory; mathematical morphology; Brazilian hydro-electric plant; IV-FMAM models; complete lattices; computational simplicity; fuzzy morphological associative memories; general type-2 fuzzy sets; interval-valued fuzzy associative memories; interval-valued fuzzy set theory; interval-valued type-2 FMAM; mathematical foundations; mathematical morphology; mathematical simplicity; representable conjunctions; streamflow prediction; type-2 fuzzy systems; Associative memory; Autoregressive processes; Equations; Fuzzy sets; Lattices; Mathematical model; Predictive models;

fLanguage

English

Publisher

ieee

Conference_Titel

IFSA World Congress and NAFIPS Annual Meeting (IFSA/NAFIPS), 2013 Joint

Conference_Location

Edmonton, AB

Type

conf

DOI

10.1109/IFSA-NAFIPS.2013.6608424

Filename

6608424