尚未登入 請由此登入
 
 
 










知識分享平台Eshare檢視資訊View
返回前一頁Back
      [檢舉]

標題Title: A support vector regression based prediction model of affective responses for product form design
作者Authors: 楊智傑,Meng-Dar..等
上傳單位Department: 多媒體與電腦娛樂科學系
上傳時間Date: 2009-11-16
上傳者Author: 楊智傑
審核單位Department: 多媒體與電腦娛樂科學系
審核老師Teacher: 楊智傑
檔案類型Categories: 論文Thesis
關鍵詞Keyword: Kansei engineering, Product form design, Support vector regression, Genetic algorithm, Neural network
摘要Abstract: In this paper, a state-of-the-art machine learning approach known as support vector regression (SVR) is introduced to develop a model that predicts affective responses (ARs) of consumers. Mobile phone design was selected to demonstrate the proposed methodology. First, pairwise adjectives were used to describe the ARs toward product samples. Second, the product form features (PFFs) of mobile phones were examined systematically and then stored either as continuous or discrete attributes. The adjective evaluation data of consumers were gathered from questionnaires. Finally, prediction models based on different adjectives were constructed using SVR, which trained a series of PFFs and the average AR rating of all the respondents. The performance of two kernel functions, including polynomial and RBF kernel, were compared. The real-coded genetic algorithm (RCGA) was used to determine the optimal training parameters of each prediction model using a different kernel function. These models are capable of accurately predicting the ARs of consumers from a new set of product samples.

檔案名稱
FileName
檔案大小
Size
檔案格式
Format
瀏覽次數
Browses
下載次數
Downloads
2010_8_0782fd93.pdf 262Kb pdf 849 524
文件中檔案:
 

開啟檔案Download
 
 
返回前一頁