| 研究生: |
徐堤 Hsu, Ti |
|---|---|
| 論文名稱: |
運用普氏分析與支撐向量機建構產品外形特徵與產品意象之關聯 Building the Relationship between Product Form Features and Product Image Using Procrustes Analysis and Support Vector Machine |
| 指導教授: |
謝孟達
Shieh, Meng-Dar |
| 學位類別: |
碩士 Master |
| 系所名稱: |
規劃與設計學院 - 工業設計學系 Department of Industrial Design |
| 論文出版年: | 2009 |
| 畢業學年度: | 97 |
| 語文別: | 中文 |
| 論文頁數: | 133 |
| 中文關鍵詞: | 造形特徵 、因素分析 、普氏分析 、數位相機 、支撐向量機 |
| 外文關鍵詞: | Support Vector Machine, Form features, Factor analysis, Digital camera, Procrustes analysis |
| 相關次數: | 點閱:79 下載:2 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
產品快速發展的今日,成功的產品設計,設計者是否能夠滿足消費者的需求與了解產品意象是非常關鍵的。而產品的外形特徵對於產品意象有著相當重要的影響。本研究使用感性工學的方法,並以數位相機為例,利用語意差異實驗,針對一群具代表性的產品樣本來記錄消費者的情感反應。第一階段以因素分析(factor analysis)在最初的感性語彙中取出重要性較低的因素,其後以普氏分析(Procrustes analysis)在每個步驟中判斷資訊損失量(RSSDs)以決定相對重要的形容詞語彙,得到五個具代表性的形容詞語彙:粗獷的、圓潤的、專業的、簡潔的、輕薄的。第二階段以造形參數化形態來拆解數位相機外形,五個形容詞語彙來表示消費者對產品的偏好程度。最後使用支撐向量機(Support Vector Machine, SVM),建構預測產品意象的預測模型,分別使用多項式核心函數與高斯核心函數,而高斯核心函數以91.48%的平均正確率優於多項式核心函數的75.40%。
Nowadays in a quick-develpoed product, a successful product design, whether designers can match consumers’requirement and understand product image is very important. Products from features have a very great effect on product image. In this study, proposes a Kansei engineering approach to research a case study of digital camera design. A semantic differential experiment asks consumers their affective responses toward a set of representative product samples. Section 1 uses factor analysis to extract underlying latent factors using an initial set of affective dimensions. The later process based on Procrustes analysis is capable of determining the relative importance of adjectives in each step according to the calculated residual sum of squared differences (RSSDs). Then got five representative affective adjectives: brutal, mellow, professional, simple, handy. Section 2 uses adjectives to describe the product images of product samples and five linguistic labels are used to linguistically evaluate the ratings toward product samples. Finally , Support Vector Machine (SVM) used to build a prediction model with a given discrimination. Polynomial kernel and Gaussian kernel applied to SVM kernel function respectively. Gaussian kernel performed very well with an average accuracy rate of 91.48%, while that of Polynomial kernel was better than 75.40%.
英文部份:
Barzilay, O. and Brailovsky V. L., On domain knowledge and feature selection using a supportvector machine, Pattern Recognition Letters, Elsevier Science Inc. 20: 475-484, 1999.
Cao, L., Support vector machines experts for time series forecasting, Neurocomputing 51: 321-339, 2003.
Chen, Y. W. and Lin C. J., Combining SVMs with various feature selection strategies, Feature Extraction, Foundations and Applications, Springer-Verlag, 273-282, 2005.
Chen, C.H., Khoo, L.P., and Yan, W., An investigation into affective design using sorting technique and Kohonen self-organising map, Advances in Engineering Software, 37(5): p. 334-349, 2006.
Han, S. H. and Kim, J., A comparison of screening methods: selecting important design variables for modeling product usability, International Journal of Industrial Ergonomics 32: 189-198, 2003.
Harada, A., The Parallel Design Methodology in the KANSEI Engineering, Report of Modeling the Evaluation Structure of Kansei, pp.309-316, 1998.
Llinares, C. and A. Page, Application of product differential semantics to quantify purchaser perceptions in housing assessment, Building and Environment, 42(7): p. 2488-2497, 2007.
Lin, Y. C., Lai, H. H., and Yeh, C. H., Consumer-oriented product form design based on fuzzy logic: A case study of mobile phones, International Journal of Industrial Ergonomics, 37(6): p. 531-543, 2007.
Osgood, C. E., The nature and measurement of meaning. Psychol Bull, 49(3), 197-237, 1952.
Kreuzbauer Robert and Malter Alan J., Changing Product Form to Influence Brand Categorization, Product Development & Management Association, 22:165–176, 2005.
Shieh, M. D. and Yang, C. C., Multiclass SVM-RFE for product form feature selection, Expert Systems with Applications, 35(1-2), pp. 531-541, 2008.
Shieh, M. D. and Yang, C. C., Classification model for product form design using fuzzy support vector machines, Computers & Industrial Engineering, 55(1), pp. 150-164, 2008.
Sahmer, K., Qannari, E.M, Procedures for the selection of a subset of attributes in sensory profiling. Food Quality and Preference. 19, 141-145, 2008.
Smits, G.F. & Jordaan, E.M., Improved SVM regression using mixtures of kernels, Proceedings of IJCNN’02 on Neural Networks, Hawaii, 12-17 May 2002.
Tanoue, C., Ishizaka, K., Nagamachi, M., Kansei Engineering: A study on
perception of vehicle interior image, International Journal of Industrial Ergonomics, 19, pp. 115-128, 1997.
Terauchi, F., Kubo, M. and Aoki, H., Development of the Synthetic Vibration Model of Human, Report of Modeling the Evaluation Structure of Kansei, pp. 263-270, 1999.
Wang, W., Z. Xu, et al, Determination of the spread parameter in the Gaussian kernel for classification and regression, Neurocomputing 55:643-663, 2003.
中文部份:
何明泉,文化商品之設計方法與策略研究,1997。
李俊宏、李柏毅、徐豐智,一個以Support Vector Machines 為主之中文文件自動分類系統的建構與特徵選取策略之分析,國立高雄應用科技大學電機工程學系,工程科技與教育學刊第二卷第一期,頁67~89,2005。
林大仁,運用支撐向量回歸與倒傳遞類神經網路建構產品造形設計之預測模式,國立成功大學工業設計學系碩士論文,2007。
周君瑞,複合感性意象之塑造-以造形特徵為基礎,國立成功大學工業設計學系碩士論文,2003。
莊盈祺,複合感性意象下產品造形建構,國立成功大學工業設計學系碩士論文,2002。
陳世和,以模組化設計與市場區隔發展產品族之研究-以自行車產業為例,國立成功大學工業設計學系碩士論文,2006。
陳志瑋,應用感性工學及田口方法於品質機能展開之發展,國立成功大學工業設計學系碩士論文,2001。
陳怡貞,消費者採用行為與意象感受程度之關聯性研究-以數位相機為例,國立成功大學工業設計學系碩士論文,2002。
黃敏菁,支撐向量機在財務時間序列預測之應用,輔仁大學金融研究所碩士論文,2004。
梅宜冬,從感性觀點探討數位產品造形意象特質-以數位攝影機為例,大葉大學設計研究所碩士論文,2004。
張建成譯,John Chris Jones 著,設計方法,六和出版,1992。
張建成、吳俊杰、劉淑君,系列化產品造形風格與設計手法研究-以OLYMPUS數位相機為例,華梵大學工業設計研究所,設計學報第12卷第3期,2007。
張嘉萍,產品造形特徵與品牌形象之一致性研究─以液晶電視為例,大同大學工業設計研究所碩士論文,2005。
楊智傑,感性工學系統之變數篩選研究,國立成功大學工業設計學系博士論文,2008。
劉恒,台灣數位相機市場消費者行為之研究,大葉大學事業經營研究所碩士論文,2002。
蔡子瑋,產品意象語言研究,國立成功大學工業設計學系碩士論文,1994。
鄭凱予,薄型電視按鍵介面與形態構成要素之研究,大同大學工業設計研究所碩士論文,2006。
鄭昭明,認知心理學-理論與實踐,桂冠出版事業股份有限公司,1993。
魏士超,應用網際網路建立產品造形意象設計系統之研究,國立成功大學工業設計學系碩士論文,2000。