Title :
Comparative analysis of two fuzzy rule base optimization methods
Author :
Johanyák, Z.C. ; Papp, O.
Author_Institution :
Inst. of Inf. Technol., Kecskemet Coll., Kecskemet, Hungary
Abstract :
Rule base optimization is a key step in fuzzy model identification that determines the performance of the fuzzy system. In this paper, we examine two optimization solutions, i.e. the cross-entropy method and a hill climbing approach based heuristic method. They are used and compared in case of four benchmarking problems. In each case the initial rule base is created by a fuzzy clustering based method.
Keywords :
fuzzy systems; knowledge based systems; optimisation; cross-entropy method; fuzzy clustering based method; fuzzy model identification; fuzzy system; heuristic method; hill climbing approach; rule base optimization; Benchmark testing; Fuzzy sets; Fuzzy systems; Optimization methods; Performance analysis;
Conference_Titel :
Applied Computational Intelligence and Informatics (SACI), 2011 6th IEEE International Symposium on
Conference_Location :
Timisoara
Print_ISBN :
978-1-4244-9108-7
DOI :
10.1109/SACI.2011.5873006
