A study on the color and style collocation of mobile phones using neural network method

Shing Sheng Guan, Yang Cheng Lin

Research output: Contribution to journalArticlepeer-review

12 Citations (Scopus)


There are a variety of mobile phones in the market. In order to attract consumers, the manufacturers of mobile phones vary the designs in style and in color. However, what style of mobile phone is just suitable for the color? If the color change of mobile phone could influence the style of mobile phone. In order to explore the relationships among color, style and Kansei vocabularies, thirty popular mobile phones were used and investigated. The liner model of Quantity and the non-liner model of Neural Network are used to analyze the experimental results. Finally, the non-liner model of Neural Network is used to infer the styles and colors of new mobile phones. This study showed that: 1) From results of the liner model of Quantitative Theory Type I, the color factor is more important than the style factor on the Kansei vocabularies in this case. 2) The “Value” factor primarily affects the “single-complex” and “leisure- formal” Kansei vocabularies, but the “Hue” factor mainly influence the “smart-rustic” Kansei vocabularies. 3) The “Chroma” is the least significant factor on any pair of Kansei vocabularies among Hue, Chroma and Value. 4) According to the results of testing new samples, the Neural Network has the highest predicted accuracy on the “single-complex” Kansei vocabularies; followed by the “smart-rustic” Kansei vocabularies. However, it is chaotic on the “leisure-normal” Kansei vocabularies. The results of this study could be derived a Decision Support System for evaluating the relationships among the style, color and Kansei vocabularies. Finally, the methods of this study can be applied at studying the relative products.

Original languageEnglish
Pages (from-to)84-94
Number of pages11
JournalJournal of the Chinese Institute of Industrial Engineers
Issue number6
Publication statusPublished - 2001 Jan 1

All Science Journal Classification (ASJC) codes

  • Industrial and Manufacturing Engineering

Fingerprint Dive into the research topics of 'A study on the color and style collocation of mobile phones using neural network method'. Together they form a unique fingerprint.

Cite this