碩士 / 東吳大學 / 資訊管理學系 / 102 / One of the communicational ways to express people's emotions is to give presents to one another. When people are in love, they often convey their feelings to their beloved ones via gift-giving. According to the National Retail Federation, the global market of gift-giving on Valentine's Day is estimated up to US 17.6 billion dollar. As a result, there are overwhelmingly various products that people can choose as a gift. This paper aims to develop a recommendation choice of gift-giving which is suitable for college couples. This research collects different kinds of male and female lifestyles, and analyzes these data via logistic regression, rough sets and decision tree models. Then, the research compares with the three empirical results to determine the best one as the recommendation.
The totally number of valid questionnaires is 237, including 113 male and 124 female. According to the empirical results of the comparisons, the decision tree model has the highest prediction accuracy and up to 64.1%. The result of the decision tree model shows that the lifestyle "enjoy learning new knowledge and introverting people" is the most critical factors for the gift-giving recommendation. If this factor belongs to the smaller objects, people will be classified into the fashionable gifts, and the corresponding accuracy rate up to 66.7%. The second key factor is the lifestyle "join campus organizations".
The results of this paper can help firms/people solving the problem of developing/choosing gifts for couples and to reduce the time cost of finding the right gift. In addition, for businessmen, they can understand more about the potential customers of their own products so that they can make appropriate marketing strategies for the gift market.
| Identifer | oai:union.ndltd.org:TW/102SCU00396024 |
| Date | January 2014 |
| Creators | HSIEH HSIAO-HUNG, 謝孝鴻 |
| Contributors | HUANG JIH-CHENG, 黃日鉦 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 58 |
Page generated in 0.0013 seconds