查詢結果分析
相關文獻
- 應用類神經網路於產品感性設計之造形要素評比--以PDA智慧型手機為例
- 感性設計支援資訊系統之研發與應用--以個人數位助理為例
- 灰關聯分析於產品感性設計輔助造形要素評比之研究
- Neural Network Procedures for Taguchi's Dynamic Problems
- A Fast and Efficient Competitive Learning Design Algorithm Based on Weight Vector Training in Transform Domain
- 專家系統振動訊號圖型判別之研究
- 反傳遞模糊類神經網路於流量推估之應用
- 類神經網路(Neural Networks)的種類及其在影像處理上的應用
- C++Fuzzy類神經網路物件導向發展系統之建立
- 臺灣汽保費率之估計--對數線性費率模式與類神經網路之比較
頁籤選單縮合
題名 | 應用類神經網路於產品感性設計之造形要素評比--以PDA智慧型手機為例=Applying Artificial Neural Networks to Rating Form Elements of Product Affective Design--Using PDA Smart Phone as an Example |
---|---|
作者 | 王振琤; | 書刊名 | 樹德科技大學學報 |
卷期 | 14:1 2012.01[民101.01] |
頁次 | 頁243-262 |
分類號 | 440.8 |
關鍵詞 | 類神經網路; 感性設計; 產品意象; 造形要素; 重要性評比; Artificial neural network; Affective design; Product image; Form element; Importance rating; |
語文 | 中文(Chinese) |
中文摘要 | 產品的外觀吸引力所產生使用者的愉悅性與滿意度,無疑地是產品成功的關鍵因素之一,因此更效的造形要素重要性評比程序是感性設計中不可或缺之機制。以往應用類神經網路建立非線性的感性設計模式,探討產品意象與造形要素間之關係,雖然能建立精度較高的感性模式與進行特定造形要素組合之意象預測,然而如何對輸入變數造形要素的評比方法仍未論述完善,因此本研究主旨在於提出一種能依據感性模式進行對造形要素的評比方法,釐清哪些造形要素較能符合產品意象的需求,以利遂行產品之感性設計。 為更效進行造形要素之評比,施行程序劃分為4個階段進行研究:(1) 前置性研究,確認32個目標產品之實驗樣本、確認9項造形要素與32種類目與意象語彙對,並進行樣本於「方便的-困難的」感性評價實驗,建立形態要素評量矩陣;(2) 應用非線性類神經網路發展BPNN感性設計支援資訊系統,以解析與建立產品感性設計模式,並進行殘差分析與信賴度t檢定以驗證該模式的更效性,結果顯示預測值與實際測量值之差異不具備顯著性,因此該感性模式具備可行性;(3) 利用已完成之網路學習結果,以Garson演算法進行類神經網路之輸入變數權重分析,獲得輸入變數 (造形要素) 對應於輸出參數 (產品意象) 之相對重要性,造形要素之權值評比結果重要度最高者為方向鍵形式 (FE5) ,最低者為頂端造形 (FE2) ,得到消費者對產品意象於造形上的偏好結構;(4) 應用已建立的BPNN感性設計模式,可找出滿足特定產品意象之最佳造形要素組合。本研究所提之非線性感性設計模式與造形要素評比機制,可更效協助設計者釐清造形要素的優先度,此外設計者可以透過最佳組合形態因子,以增強特定的產品意象方式建立產品造形風格標竿,在此設計概念化下可豐富與最佳化產品多樣性。 |
英文摘要 | A significant factor to a successful product is its appearance attraction which can evoke users‟ pleasure and satisfactions. Hence, an effective importance rating mechanism of form elements is indispensable in product affective design. Formerly, the use of artificial neural network is to establish nonlinear models for exploring the relationships between product images and form elements. The merit of obtained model usually has precise capability to forecasting image for a given form element combination. However, the rating method of input variables is still not addressed well. Hence, this study aims to propose a form element rating method based on the obtained affective design model. This proposed method could assist designers clarifying the critical form elements which most influences product image. To perform an effective form element rating, the implementation procedure is composed of four stages: (1) Conduct a preliminary study that identified 32 experimental samples, nine form elements and 32 categories, and image word pairs. We conducted an affective evaluating experiment of “convenient-difficult” image to obtain morphological element evaluation matrix. (2) Use nonlinear artificial neural network to develop a BPNN design support information system to analyze and establish product affective design model. Residue analysis and statistical t-test were conducted to verify model validity. The results shows that no significant difference between the predicted values and practical values. (3) Use Garson algorithms to determine the weighting functions of input variables based on the obtained trained results of neural network. The results shows that the form element “arrow key type” (FE5) is the most critical, but the “top shape” (FE2) the least critical, thus we can obtain consumers‟ preference structure with relation to form elements for a given product image. (4) Use obtained BPNN affective design model to find out the best combinations for form elements for a specific given product image. The proposed nonlinear affective design model and form element rating methodology could assist designers elucidating the priorities of form elements. In addition, designers can create product form benchmark by enhancing specific product image through the best form element combinations. As thus design conceptualization, we can enrich and optimize product variety. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。