Kansei knowledge extraction based on evolutionary genetic algorithm: an application to e-commerce web appearance design |
| |
Authors: | Qing-Xing Qu |
| |
Affiliation: | 1. Department of Management Science and Engineering, School of Business Administration, Northeastern University, Shenyang 110819, P.R. Chinayantaiquqingxing@163.com |
| |
Abstract: | Considering the overall consumer preference based on Kansei engineering, this paper focuses on optimising the appearance design of e-commerce web. In the beginning, we have used iView X RED Eye Tracking Systems experimental apparatus produced by SensoMotoric Instruments (SMI) in Germany to extract web design elements, and then several representative webs are designed based on orthogonal test design, following by surveys we have made. Furthermore, structural equation models are established in order to obtain a single preference factor on the influence of e-commerce web design. Finally, based on the neural networks (NNs) and evolutionary genetic algorithm approach, the global optimisation appearance design of the e-commerce web is fetched by simulation on computer, providing the effective suggestions for the e-commerce web designer. This research paper presents a systematic approach that convert consumer's Kansei knowledge into usable product multi-dimensional design variables. |
| |
Keywords: | Kansei engineering web appearance design structural equation model neural networks evolutionary genetic algorithm |
|
|