Interval fuzzy sets in the tasks of recognizing and predicting the states of complex objects in conditions of incomplete data

Abstract

In the tasks of recognizing and predicting the states of complex objects, a common situation is when it is necessary to make decisions with contradictory or incomplete data, which is caused by the presence of a certain level of “noise” or gaps in the data. To solve the existing problem, interval fuzzy sets, or fuzzy sets of type-2, are used, when the values of the membership functions are not a number, but an interval. The report shows the appearance of an interval output in fuzzy logic systems of type-2 when using interval membership functions and the construction of a set of fuzzy models of type-2 using set-theoretic operations for making effective decisions in applied problems.