| | |
Summary: Adaptation of Fuzzy Inferencing: A Survey
Payman Arabshahi, Robert J. Marks II, and Russell Reed
Department of Electrical Engineering
University of Washington FT-10
Seattle, WA 98195 USA
Abstract-- Fuzzy inference has numerous applications, rang-
ing from control to forecasting. A number of researchers
have suggested how such systems can be tuned during ap-
plication to enhance inference performance. Inference pa-
rameters that can be tuned include the central tendency and
dispersion of the input and output fuzzy membership func-
tions, the rule base, the cardinality of the fuzzy membership
function sets, the shapes of the membership functions and
the parameters of the fuzzy AND and OR operations. In
this paper, an overview of these tuning procedures is given.
An extensive bibliography is provided of recent literature on
the topic.
I. INTRODUCTION
A general fuzzy inference system consists of three parts (see
Fig. 1). A crisp input is fuzzified by input membership functions
|
