A Genetic Fuzzy System with Inconsistent Rule Removal and Decision Tree Initialization

This paper presents a genetic fuzzy system for identification of Paretooptimal Mamdani fuzzy models (FMs) for function estimation problems. The method simultaneously optimizes the parameters of fuzzy sets and selects rules and rule conditions. Selection of rules and rule conditions does not rely only on genetic operators, but it is aided by heuristic rule and rule conditions removal. Instead of initializing the population by commonly usedWang-Mendel algorithm, we propose a modification to decision tree initialization. Experimental results reveal that our FMs are more accurate and consist of less rules and rule conditions than the FMs obtained by two recently published genetic fuzzy systems [2, 3].

[1]  Marco Laumanns,et al.  SPEA2: Improving the strength pareto evolutionary algorithm , 2001 .

[2]  Francisco Herrera,et al.  Ten years of genetic fuzzy systems: current framework and new trends , 2004, Fuzzy Sets Syst..

[3]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[4]  Z. Xing,et al.  On Generating Fuzzy Systems based on Pareto Multi-objective Cooperative Coevolutionary Algorithm , 2007 .

[5]  Francisco Herrera,et al.  Solving Electrical Distribution Problems Using Hybrid Evolutionary Data Analysis Techniques , 2004, Applied Intelligence.

[6]  Hannu Koivisto,et al.  Identification of interpretable and accurate fuzzy classifiers and function estimators with hybrid methods , 2007, Appl. Soft Comput..

[7]  Hisao Ishibuchi,et al.  Effect of rule weights in fuzzy rule-based classification systems , 2001, IEEE Trans. Fuzzy Syst..

[8]  Uzay Kaymak,et al.  Similarity measures in fuzzy rule base simplification , 1998, IEEE Trans. Syst. Man Cybern. Part B.

[9]  Hannu Koivisto,et al.  Detection of Safe and Harmful Bioaerosols by Means of Fuzzy Classifiers , 2008 .

[10]  Beatrice Lazzerini,et al.  A Pareto-based multi-objective evolutionary approach to the identification of Mamdani fuzzy systems , 2007, Soft Comput..

[11]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[12]  Hannu Koivisto,et al.  Fuzzy classifier identification using decision tree and multiobjective evolutionary algorithms , 2008, Int. J. Approx. Reason..

[13]  Kim-Fung Man,et al.  Multi-objective hierarchical genetic algorithm for interpretable fuzzy rule-based knowledge extraction , 2005, Fuzzy Sets Syst..

[14]  Francisco Herrera,et al.  A Multi-Objective Evolutionary Algorithm for Rule Selection and Tuning on Fuzzy Rule-Based Systems , 2007, 2007 IEEE International Fuzzy Systems Conference.

[15]  Jerry M. Mendel,et al.  Generating fuzzy rules by learning from examples , 1992, IEEE Trans. Syst. Man Cybern..

[16]  J. M. Edmunds,et al.  On fuzzy logic controllers , 1991 .

[17]  Ferenc Szeifert,et al.  Data-driven generation of compact, accurate, and linguistically sound fuzzy classifiers based on a decision-tree initialization , 2003, Int. J. Approx. Reason..