A Review on Integrating Evolutionary Multiobjective Optimization in Developing Fuzzy Systems

Fuzzy systems are developed to model the information inexactness and uncertainty in the real world problems. The development of fuzzy system can be considered as an optimization task. So Evolutionary Algorithms are well utilized to design fuzzy systems. The fuzzy systems have two strong features tha...

Full description

Saved in:
Bibliographic Details
Published in2018 International Conference on Computational and Characterization Techniques in Engineering & Sciences (CCTES) pp. 252 - 257
Main Authors Khattri, Vipin, Nayak, Sandeep Kumar
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.09.2018
Subjects
Evolutionary Algorithm(EA)
Evolutionary computation
Evolutionary Multiobjective Optimization(EMO)
Fuzzy System
Fuzzy systems
Genetic Algorithm(GA)
Genetic algorithms
Genetic Fuzzy System(GFS)
Interpretability-Accuracy Trade-off
Multiobjective Evolutionary Algorithm(MOEA)
Pareto Front
Pareto optimization
Sociology
Online AccessGet full text

Cover

More Information
Summary:Fuzzy systems are developed to model the information inexactness and uncertainty in the real world problems. The development of fuzzy system can be considered as an optimization task. So Evolutionary Algorithms are well utilized to design fuzzy systems. The fuzzy systems have two strong features that must be considered during the development i.e. interpretability and accuracy. Interpretability is the subjective feature and not easy to quantify and accuracy shows the closeness between the real and modeled systems. These features are contradicting with each other and lead to interpretability-accuracy trade-off. This is basically combined as a multiobjective optimization problem and further it is solved by the multiobjective optimization evolutionary algorithm (MOEA). This is the reviews the approaches developed under these issues. The future of this research line is also discussed.
DOI:10.1109/CCTES.2018.8674108