A New Design of Mamdani Complex Fuzzy Inference System for Multiattribute Decision Making Problems

This article proposes the Mamdani complex fuzzy inference system (Mamdani CFIS) to improve performance of the classical FIS and complex FIS. The applicability of the proposed CFIS is demonstrated by applying it to six commonly available datasets from UCI Machine Learning under the comparison with Ma...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on fuzzy systems 2021-04, Vol.29 (4), p.716-730
Main Authors: Selvachandran, Ganeshsree, Quek, Shio Gai, Lan, Luong Thi Hong, Son, Le Hoang, Giang, Nguyen Long, Ding, Weiping, Abdel-Basset, Mohamed, de Albuquerque, Victor Hugo C.
Format: Article
Language:English
Subjects:
Adaptive systems
Complex fuzzy inference system (CFIS)
complex fuzzy logic (CFL)
Complex numbers
Computational efficiency
Computational modeling
Decision making
Fluctuations
Fuzzy logic
Fuzzy sets
Inference
Information science
Machine learning
Mamdani fuzzy inference system (FIS)
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This article proposes the Mamdani complex fuzzy inference system (Mamdani CFIS) to improve performance of the classical FIS and complex FIS. The applicability of the proposed CFIS is demonstrated by applying it to six commonly available datasets from UCI Machine Learning under the comparison with Mamdani FIS and the Adaptive Neuro Complex Fuzzy Inference System (ANCFIS). It is successfully proven that the proposed Mamdani CFIS is computationally less expensive and presents a more efficient method to handle time-series data and time-periodic phenomena, among all the fuzzy IS found thus far in the literature. Furthermore, the novelty of CFIS mainly lies in its implementation of the complex number throughout the entire procedures of computation. This gives much greater flexibility of implementing unexpected, nonlinear fluctuations.
ISSN:1063-6706
1941-0034
DOI:10.1109/TFUZZ.2019.2961350