Please use this identifier to cite or link to this item:
Title: Generalized Fuzzy Entropy Optimization Methods for Fuzzy Data Analysis
Authors: İnce, Nihal
Keywords: Fuzzy entropy measure
Minimum cross fuzzy entropy principle
Simulation of fuzzy random variables
Issue Date: 2022
Publisher: Springer Science and Business Media Deutschland GmbH
Abstract: The fuzzy entropy is used to express the mathematical values of the fuzziness of fuzzy sets and is defined by using the concept of membership function. The maximum entropy principle attempts to choose the membership function with a finite number of the fuzzy values subject to constraints generated by given moment vector functions that have maximum entropy value. On the other hand, the minimum cross-entropy principle tells us that out of all membership functions satisfying given moment constraints, select the one that is closest to the given a priori membership function. This study is connected with new Generalized Minimum Cross Fuzzy Entropy Methods (GMinx(F)EntM) in the form of MinMinx(F)Ent and MaxMinx(F)Ent methods. The aim of this study consists of applying GMinx(F)EntM on given simulated fuzzy data. The performances of distributions (MinMinx(F)Ent)m and (MaxMinx(F)Ent)m are compared by Chi-Square, Root Mean Square Error, coefficient of determination criteria and fuzzy cross entropy measure. The obtained results show that Generalized Minimum Cross Fuzzy Entropy Optimization distributions give significant results in the data modeling for fuzzy data analysis. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Description: International Conference on Intelligent and Fuzzy Systems, INFUS 2021 -- 24 August 2021 through 26 August 2021 -- -- 264409
ISBN: 9.78E+12
ISSN: 2367-3370
Appears in Collections:Scopus İndeksli Yayınlar Koleksiyonu
Ulaştırma Meslek Yüksekokulu Yayın Koleksiyonu

Show full item record

CORE Recommender

Page view(s)

checked on Oct 3, 2022

Google ScholarTM



Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.