DocumentCode :
1957426
Title :
The linguistic gradient of a fuzzy logic rule-base
Author :
Dick, Scott ; Kandel, Abraham
Author_Institution :
Dept. of Comput. Sci. & Eng., Univ. of South Florida, Tampa, FL, USA
Volume :
2
fYear :
2000
fDate :
2000
Firstpage :
911
Abstract :
We introduce the linguistic gradient operator, a novel concept in fuzzy systems theory. A multiple-input, single-output fuzzy rule-base of the Mamdani type is modeled in a linguistic state space. The consequents of the rules are treated as a function, whose domain is the linguistic state space representing the rule antecedents. The linguistic gradient operator operates on the consequent function, and produces an n-dimensional vector pointing in the direction of the greatest local increase in the rule consequents. Thus, the linguistic gradient shows how a given rule relates to its neighbors in linguistic space. An example demonstrates that the linguistic gradient vector is a linguistic analogue of the classical gradient vector
Keywords :
fuzzy control; fuzzy logic; fuzzy set theory; gradient methods; knowledge based systems; linguistics; state-space methods; Mamdani type; fuzzy control; fuzzy logic; gradient vector; linguistic gradient; linguistic state space; rule consequents; rule-base; Calculus; Cellular neural networks; Control systems; Fuzzy control; Fuzzy logic; Fuzzy systems; State-space methods;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Fuzzy Systems, 2000. FUZZ IEEE 2000. The Ninth IEEE International Conference on
Conference_Location :
San Antonio, TX
ISSN :
1098-7584
Print_ISBN :
0-7803-5877-5
Type :
conf
DOI :
10.1109/FUZZY.2000.839153
Filename :
839153
Link To Document :
بازگشت