Title :
The Effect of Different Fuzzy Partition Parameterization Strategies in Gradient Descent Parameter Identification
Author :
Johanyák, Zsolt Csaba ; Kovács, Szilveszter
Author_Institution :
Inst. of Inf. Technol., Kecskemet Coll., Kecskemet
fDate :
Yearly 17 2007-May 18 2007
Abstract :
One of the most critical steps during the development of a fuzzy system is the identification of the fuzzy rule base and the fuzzy partitions, the so-called "tuning". This paper intends to present a comparative study of three different fuzzy partition parameter identification methods with respect to the effect of different fuzzy partition parameterization strategies.
Keywords :
fuzzy set theory; fuzzy systems; gradient methods; knowledge based systems; parameter estimation; fuzzy partition parameterization; fuzzy rule base; fuzzy system; gradient descent; parameter identification; Clocks; Educational institutions; Fuzzy sets; Fuzzy systems; Ice; Information technology; Interpolation; Parameter estimation; Partitioning algorithms; Performance analysis; fuzzy rule interpolation; fuzzy system tuning; rule base optimization; sparse rule base;
Conference_Titel :
Applied Computational Intelligence and Informatics, 2007. SACI '07. 4th International Symposium on
Conference_Location :
Timisoara
Print_ISBN :
1-4244-1234-X
DOI :
10.1109/SACI.2007.375499