Kansei evaluation for group of users: A data-driven approach using dominance-based rough sets

摘要

Kansei refers to people's subjective feeling and impression. Kansei evaluation devotes to assessing users' preferences for product items according to multiple Kansei attributes, thus supporting the decision making of consumers and/or designers. The objective of this paper is to propose a data-driven approach for addressing user group oriented Kansei evaluation. The approach consists of three phases. The first phase identifies the representative Kansei attributes and product samples of the product domain to gather exemplary evaluation dataset from sampled representative users. In light of the specified Kansei need and relying on the dominance-based rough set approach, the second phase constructs the collective decision table so as to further infer the collective preferential information in terms of dominance-based decision rules and Kansei importance weights. The third phase presents a two-step sequential heuristic model for characterizing users' affective preference behavior: (1) a multicriteria classifier using dominance-based decision rules for product sorting, partly simulating the satisflcing heuristic; and (2) a simple choice strategy for product ranking, manifesting the CONF heuristic. A case study involving the toaster domain was conducted to verify the proposed approach. The theoretical and practical implications of the proposed approach are also discussed.