A type-2 fuzzy concepts lexicalized representation by perceptual reasoning and linguistic weighted average: A comparative study

Abstract

Concepts lexicalized in natural language are uncertain. Actually, there are cooperation between type-1 fuzzy sets (T-1 FSs) and the psychology of concepts for manipulating knowledge. Our approach shows that concepts can be equalized to interval type-2 fuzzy sets (IT-2 FSs) by using a Computing With Words (CWW) model. CWW is a theory that passes from computing with crisp values or measurements to CWW or concepts. This paper presents a comparative study between the Perceptual Reasoning (PR) and the Linguistic Weighted Average (LWA) and implements them using a mammography database. These two approaches are implemented in the CWW engine of a CWW model; they characterize linguistic uncertainties existing in concepts by using IT-2 FSs. The results obtained demonstrate that the PR approach give results similar to concepts in the code-book. This paper insists on the fact that concepts can be represented by IT-2 FSs in the psychology of concepts.