Genetic tuning for improving Wang and Mendel's fuzzy database

Authors:
E. R. R. Kato;O. Morandin;M. Sgavioli;B. D. Muniz
Affiliations:
Department of Computer Science, Federal University of São Carlos, São Carlos, Brasil;Department of Computer Science, Federal University of São Carlos, São Carlos, Brasil;Department of Computer Science, Federal University of São Carlos, São Carlos, Brasil;Department of Computer Science, Federal University of São Carlos, São Carlos, Brasil
Venue:
SMC'09 Proceedings of the 2009 IEEE international conference on Systems, Man and Cybernetics
Year:
2009

Citing 6
Cited 0

Fuzzy sets and applications

Fuzzy sets and applications
Fuzzy sets and fuzzy logic: theory and applications

Fuzzy sets and fuzzy logic: theory and applications
Tackling Real-Coded Genetic Algorithms: Operators and Tools for Behavioural Analysis

Artificial Intelligence Review
Neuro-fuzzy systems for function approximation

Fuzzy Sets and Systems - Special issue on analytical and structural considerations in fuzzy modeling
COR: a methodology to improve ad hoc data-driven linguistic rulelearning methods by inducing cooperation among rules

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
A Proposal for the Genetic Lateral Tuning of Linguistic Fuzzy Systems and Its Interaction With Rule Selection

IEEE Transactions on Fuzzy Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper aims to use the Genetic Tuning, an evolutionary approach to improving an existing fuzzy database of a fuzzy system, which the rule base is generated by example, through fitness function manipulation. Evolutionary Computing is shown efficient for solving problems where the solution space is too large and its complete analyses is computationally unfeasible. The evolutionary solution will be applied in the adjustment of fuzzy sets generated by the Wang and Mendel's method. Wang and Mendel's method high performance has been clearly demonstrated. Although, inducing genetic tuning, a better set of parameters of the database can be achieved. Our proposal is to find a better representation of the input data attributes and find a minimum distance between the original output of the data set and the output given by the Wang and Mendel's method keeping a good interpretability of the sets and improving the accuracy of the system. To analyze the performance we will use tree different data set, considering Real Coding Scheme to chromosome codification.