Membership function based on similarity and dissimilarity

Abstract

Deciding membership function of an object to uncertain class of objects in an information system is essential particularly, all things considered, displaying issues. This paper will present a comparison between different ways of constructing membership function of an object. We initiate a concept based on dissimilarity and similarity between objects using values of their features. The fundamental tool is classes resulted from dissimilarity and similarity. We will also express a classification relation and minimum and maximum memberships correspond to these classifications.

Keywords

• rough membership
• dissimilarity measures and similarity