The Application of Data Mining for Building Personalized Recommender Systems-Base on Hokkaido,Japan

• Main Authors: HSIAO,CHUN-CHIEN, 蕭竣謙
• Other Authors: YEH,HUI-CHUNG
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/x7grk4

碩士 === 國立高雄應用科技大學 === 企業管理系 === 104 === ABSTRACT Today most people have experience in tourism, with more and more people go abroad, travel agencies reaching saturation, industry competition are difficult, so just be more understanding of customer needs in order to win customer trust. Customers usually interested in how to arrange the trip, customers recommended tourist attractions itinerary is relatively important . Travel agencies effective use pre-emptive advantage to increase the customer base and profits, and find more features which customers, consumer groups, to improve the overall performance is the focus of this study. In this study, Base on Hokkaido, Japan, design-related itinerary and questionnaire, also set a problem item questionnaire scores with ratings after expert discussions, and through the application of data mining technology to do the recommended system. The process can be divided into four stages : the first step is to collect data, in a second step for the first step of the information collected to do sorting, use information previously sorted out in the third step to do data mining application, the recommended system analysis, and finally the fourth step is comparison based on questionnaires and two recommended basic information system, hoping to help travel agents can quickly and effectively to recommend customers the most suitable itinerary. Therefore, we can find two recommended in different ways, first recommended way is expert score only get counted out after personal travel points; the second is the recommended way to group data points, recommended for the public use of the second approach not only allows operators to quickly categorize customer information, in contrast with the first also more cautious.