Hybrid Association Mining and Refinement for Affective Mapping in Emotional Design

Abstract

Emotional design entails a bidirectional affective mapping process between affective needs in the customer domain and design elements in the designer domain. To leverage both affective and engineering concerns, this paper proposes a hybrid association mining and refinement (AMR) system to support affective mapping decisions. Rough set and K optimal rule discovery techniques are applied to identify hidden relations underlying forward affective mapping. A rule refinement measure is formulated in terms of affective quality. Ordinal logistic regression (OLR) is derived to model backward affective mapping. Based on conjoint analysis, a weighted OLR model is developed as a benchmark of the initial OLR model for backward refinement. A case study of truck cab interior design is presented to demonstrate the feasibility and potential of the hybrid AMR system for decision support to forward and backward affective mapping.