Mining affective words to capture customer's affective response to apparel products

作者:Zhao, Y.; Song, J.; Montazeri, Alireza; Gupta, M. M.; Lin, Y.; Wang, C.; Zhang, W. J.*
来源:Textile Research Journal, 2018, 88(12): 1426-1436.
DOI:10.1177/0040517517712092

摘要

Contemporary apparel design practice is such that the functional and ergonomic aspects can be readily rationalized and thus computerized, but this is not true for the aspects of affect or emotion. Design for emotion or affect remains an ad hoc exercise. In the apparel industry, affects or emotions of both wearers and audiences are very important. This paper presents a work with an overall objective to rationalize the affective property of apparel. To achieve this overall objective, the first step is to have a language (a set of words in this case) to describe the customer's need in the affective attribute or property of apparel into a technical specification. In the work reported in this paper, this language (simply, a set of words) has been developed by the application of a proposed data mining procedure with a proprietary tool. A preliminary experiment was performed to validate this language - how to accurately capture the voice of customer in the aspect of affects in this case. There are two contributions out of this work: (1) finding a set of words that describe the affective property of apparel to capture the voice of customer in the aspect of affect, which is a foundation for the computer-based affective design of apparel; and (2) formulation of a new data mining process for searching affective words from the internet, which has a generalized implication to affective design in other domains of products, such as furniture.