DATA PROCESSING AND ANALYSIS

Method for classification of objects with fuzzy values of features

V. G. Sinuk, S. V. Kulabukhov

Federal State Budget Educational Institution of Higher Education "Belgorod State Technological University named after V.G. Shukhov", Belgorod, Russia

Abstract: The paper is dedicated to the development of classification method for objects, which have their features represented as fuzzy sets. The method is based on computation of the compatibility of the composite premise, which defines an individual class, and the fuzzy features of the objects. The compati-bility is represented by means of fuzzy truth values. The results of defuzzification of these values are used to compare the compatibilities and thus to determine the class of the object.

Keywords: extension principle, fuzzy truth value, linguistic variable.

Language: English

DOI: 10.14357/20718632220107

Bibliographic databases:

•