Multicriterial optimization of information granules in fuzzy If-Then rules
Аннотация
Investigation of trade-off between
characteristics of information granules for construction of
fuzzy rule-based model is challenging scientific problem.
Main optimization requirement in information granules
estimation (antecedent and consequent parts of fuzzy rules)
are cardinality which characterizes justifiability (reliability)
of information granules in light of the available experimental
data and specificity that characterizes semantics of
information granules. This paper is devoted to a multicriterial
approach to solving of optimization problem of
construction of interval and fuzzy information granules.
Validity of the proposed approach is proved by numerical
examples.