The Study of Extendable Social Affiliation Network on Chance Discovery - A Case of Cell Phone Design

• Main Authors: Mu-Hua Lin, 林木花
• Other Authors: Chao-Fu Hong Ph.D.
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/93417189768447851628

碩士 === 真理大學 === 管理科學研究所 === 93 === In this paper, we apply the affiliation network distance of small world and chance of lateral marketing to find the relation of user’s data. And understand people that it is not easy to make a decision under limited environment and information. It is difficult to find the satisfied product because it is at random to crossover and mutation in traditional IEC. But we can see that some successful cases are in the Milgram's experiment. This means that as long as defined the strategy, using IEC can find the satisfying solution or objective. So we try to analyze the chance with the concept of affiliation distance and conjoint analysis in this research. Using the concept of affiliation graph interactive evolutionary computation to find the chance. And help users to find their preference. We apply AGIEC model to the cell-phone design in this research. From experiment data we know user can find their satisfying solution combines the vertical expands and horizontal expands. And proved in Kleinberg condition that people's social domain has the best result of searching when degree from two to three. To calculate the classifications weight of the cell-phone component in this research, in order to combine the cell-phone that user's preference. Proving from the result of the experiment that increasing the classification level contributes to find the satisfying solution.