A linguistic representation based approach to modelling Kansei data and its application to consumer-oriented evaluation of traditional products

This paper aims at introducing a new method for Kansei data modelling, which is basically based on the linguistic interpretation of Kansei data and the probabilistic semantics of fuzzy sets. We first interpret a semantic differential scale used in Kansei experiment as a linguistic variable and then propose a new method of how the Kansei value of products can be represented as a probability distribution on the set of Kansei linguistic labels of the linguistic variable. We further extend this Kansei data modelling approach to the cases where the Kansei value is imprecisely given as an interval or a fuzzy number. Furthermore, the proposed approach for modelling Kansei data will be used to develop a target-based model for consumer-oriented evaluation. Finally, a case study for consumer-oriented evaluation of hand-painted Kutani cups, a traditional product in Ishikawa, Japan, is conducted to show the applicability of the proposed Kansei data modelling approach.