摘要
基于消费者偏好度量法和数据统计分析法相结合的新技术路线,优化了传统主成分回归分析法的自变量筛选方法,建立起手机的外观评估预测模型和评测属性的权重-分值图,为改良现有产品和评估新产品方案提供了明确有效的科学依据.经实例验证,6名专家使用该方法与30名手机使用者对手机外观的评分结果基本一致,表明利用该方法对产品外观进行评估,可以节省很多人力和时间,从而降低产品外观评估的成本、提高产品外观评估的效率.该方法为设计人员提供一种新的切实可行的、便于操作的产品外观评估方法.
The paper presents a novel method based on psychological preference metric and statistical analysis to improve the way of selecting variables of original principal component regression. In addition, a prediction model of mobile appearance evaluation and weight-score graph of evaluated attributes are constructed which provides a science basis for improve both the current and new product appearance. Final tests show that the appearance evaluation result of 6 experts using the new method is similar to the result of 30 mobile user surveys. It shows by this method, it can save a lot of manpower and time in the process of appearance evaluation can be saved, so as to reduce the cost of product appearance evaluation and improve the assessment efficiency. Therefore, the new method can provide designers a feasible method for evaluating product appearance.
出处
《东南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2011年第4期739-743,共5页
Journal of Southeast University：Natural Science Edition
基金
航空科学基金
航空电子系统综合技术国防科技重点实验室联合资助项目(20085569014)
关键词
产品外观评估
主成分回归分析
李克特度量法
消费者偏好
product appearance evaluation
principal component regression
Liket scale
consumer preferences
