摘要
为准确把握用户的情感需求,建立产品意象和用户情感需求之间的映射关系,以电热水壶为样本进行试验,采用多维尺度分析、聚类分析等方法构建产品外观空间,并采用语义差异量表、因子分析等方法构建产品多维感性语义空间。将产品的外观空间和语义空间结合,利用支持向量回归,构建产品多维感性语义识别模型并加以验证。结果表明:依据感性语义数据所建立的支持向量机语义识别模型性能良好,可运用到实际产品的感性语义分类预测当中。
In order to grasp users emotional need accurately, the mapping relationship between product image and user emotional demand was established. The electric kettle was taken as product samples for research. The product modeling space was established using multi-dimensional scale analysis and cluster analysis. The multi-dimensional perceptual semantic space was constructed using semantic differential scale and factor analysis. The product forming space and semantic space were combined to build and validate the multi-dimensional perceptual semantic identification model using supported vector regression. The results showed that the support vector machine （SVM） model built on perceptual semantic data had a good performance, and could be applied to the perceptual semantics classification of real products.
出处
《机械设计》
CSCD
北大核心
2017年第3期105-110,共6页
Journal of Machine Design
基金
国家科技支撑计划资助项目(2015BAH21F01)
关键词
产品设计
感性工学
外观元素
支持向量机
product design
Kansei engineering
appearance elements
support vector machine
