Constructing a Travel Itinerary of High Satisfaction by Using Data Mining Techniques

• Main Authors: Yu-Hsuan Hsieh, 謝宇宣
• Other Authors: Chin-Shien Lin
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/72q48q

碩士 === 國立中興大學 === 企業管理學系所 === 107 === In recent years, Taiwan government attaches great importance to the tourism industry, many related tourism policies have been launched and also promote strategy of tourism industrial sustainable development actively. As a result, Taiwan''s tourism market continues to grow and the number of tourists has reached a new peak. However, the tourism market is constantly changing. Regardless of domestic tourism or traveling abroad, there are more and more people who prefer to choose independent travel, this has now become the new trend. In the past, relevant research did not have a specific method to provide travel itineraries that meet consumer needs. This study focused on independent travel, and uses Pearce’s travel career ladder to divide travel motivation into five constructs, combining data mining techniques to propose a travel itinerary exclusive for independent travelers. Therefore, the object of this study is Taiwanese people who visited Taichung in the past three years. The survey sample is 411. Statistical software SAS, SPSS and AMOS are used to analyze the survey samples. The results of this study show that Logit Leaf Model and Neural Leaf Model have highest accuracy, both of them exceed 85%. The empirical results not only presents the effect of combining data mining techniques but also has important management implications for tourism’s customer relationship management in practice. The accuracy of data mining is ranked from high to low as Neural Leaf Model, Logit Leaf Model, Logistic regression, Neural Networks and Decision Trees.