Measures of similarity among fuzzy concepts: A comparative analysis☆

摘要: Abstract Many measures of similarity among fuzzy sets have been proposed in the literature, and some incorporated into linguistic approximation procedures. The motivations behind these are both geometric set-theoretic. We briefly review 19 such compare their performance a behavioral experiment. For crudely categorizing pairs concepts as either “similar” or “dissimilar,” all performed well. distinguishing between degrees dissimilarity, certain were clearly superior others inferior; for few subjects, however, none distance adequately modeled judgments. Measures that account ordering on base variable proved to be more highly correlated with subjects' actual And, surprisingly, best ones focus only one “slice” membership function. Such easiest compute may provide insight way humans judge concepts.