Designing Recommendation Model With Analytic Hierarchy Process (AHP) and Image-based Presentations.

• Main Authors: Chen, Jiajun, 陳佳駿
• Other Authors: Tseng, Chinyang
• Format: Others
• Language: en_US
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/97779678162153722879

碩士 === 國立臺北大學 === 資訊管理研究所 === 100 === High accuracy of market surveys is critical because it often determines the marketing directions and strategies of a company. However, most customers are not clear to know what they really need. For company, it is hard to design or sell the appropriate products to customers if they could realize their demand. This paper applies Analytic Hierarchy Process (AHP) and image-based presentation to questionnaire design to give users instant alternative feedback with what they really need. From user decision-making behavior aspect, humans tend to make a more accurate preference decision in pair-wise factor comparison with precisely scoring to every factor. Furthermore, image-based presentations allow humans to imagine and visualize their real preference, and thus increase the accuracy and reliability of recommendations. In brief, this model focuses on improving the products’ reliability, performing real-time analysis, and increasing the purchasing intention. To evaluate the reliability of the model, the sample results pass the following tests: Cronbach’s Alpha test and pearson test. We verify all hypotheses by descriptive statistics, and T-test values of sample results show that users agree the statements of questions