發展一個LIT序列探勘方法於客製化路徑推薦系統

Developing a Location-Item-Time Sequential Pattern Mining algorithm for Personalized Route Recommendation Service

賴柏翰(Bo-Han Lai)

指導教授 ： 蔡介元

元智大學/工程學院/工業工程與管理學系/碩士(2012年)

https://doi.org/10.6838/YZU.2012.00317

摘要

近年來，遊樂園在經營上面對許多嚴酷的同業競爭，而遊樂園的經營者要如何在這樣的瞬息萬變的環境中持續經營，並提供客製化及高品質的服務給顧客，已經成為經營遊樂園中一項重要的課題。為了滿足不同需求的顧客，本研究以顧客的觀點為基礎，發展出一套路徑推薦系統，讓顧客可以知道這些被推薦的遊樂設施是在哪個區域或屬於哪個主題。為了收集顧客遊玩的順序及其遊玩時間，在遊樂園裡所有的區域都被佈滿無線射頻識別系統。本研究主要可以分為兩個階段；第一階段中，會先將收集到的資料進行前置處理，將不合理的資料進行修改，接著利用本研究提出的 LIT-PrefixSpan 演算法步驟，探勘出高頻的序列樣式。在接續的第二階段，本系統為了推薦路徑的產生，會先依照顧客輸入預計離開遊樂園的總時間，進行序列樣式的篩選，接著將篩選出的候選序列樣式與顧客輸入的遊玩區域、遊樂設施及其停留時間，利用本研究提出的相似性比對方式，藉此產生最符合顧客個人需求的遊樂園路徑規劃。最後，本研究以東京迪士尼海洋遊樂園做模擬實例，說明路徑推薦系統的完整流程，並且加入使用者偏好差異對路徑推薦的影響及系統參數分析，以增進最終路徑規劃的品質。從實驗結果顯示本研究所提出的路徑推薦系統確實可以在符合顧客輸入的限制下，給予顧客完整且多元的路徑推薦。

關鍵字

推薦系統 ； 瀏覽行為模式 ； 路徑推薦 ； 無線射頻辨識系統 ； 間隔時間序列探勘 ； 遊樂園

並列摘要

Recently, amusement parks are now facing strict challenges from other business competition. How to survive in a rapidly changing environment and provide a high quality service in terms of consumer preference has become a critical issue for amusement park managers. To satisfy the requirement, this research proposes a personalized route recommendation system that provides tourists with the facilities in which theme they should visit. In the proposed system, to detect which region a visitor is in, all regions are covered by Radio-Frequency Identification (RFID) readers. The tourist behaviors (i.e. visiting sequences and corresponding timestamps) will be collected until the tourist leaves the amusement park and stored in a route database. The proposed route recommendation system consists of two major stages. The first step is to preprocess the route sequences so that unreasonable routes are modified and corrected. After the preprocessing the route database, the second step is to discover the frequent Location-Item-Time (LIT) sequential patterns using the proposed Location-Item-Time PrefixSpan (LIT-PrefixSpan) mining procedure. In the second stage, the route suggestion procedure will filter the LIT sequential patterns under the constraints of intended-visiting time, favorite regions with its related visiting time, and favorite recreation facilities. In addition, this research develops the similarity measurement between the new user’s input vector and the candidate LIT routes from the user point of view. Finally, the route recommendation system will select the top-three suggested routes to guide the visitors. To show the feasibility of the proposed route recommendation system, the Tokyo DisneySea in Japan is used as an example. The experiments for different parameters and user preference settings are conducted to demonstrate the quality of route recommendation. Based on the experimental results, it is clear that the recommended route satisfies visitor requirement based on previous visiting experiences of tourists.

並列關鍵字

RFID ； Time-interval sequential pattern mining ； Amusement park ； Recommendation systems ； Visiting sequence ； Route suggestion

參考文獻

2. Albadvi, A., and Shabbazi M., “A hybrid recommendation technique based on product category attributes,” Expert Systems with Application, 36(9), pp. 11480-11488, 2009.

4. Chen, Y.-L, and Chiang, M.-C., and Ko, M.-T., “Discovering time-interval sequential patterns in sequence database,” Expert Systems with Applications, 25(3), pp. 343-354, 2003.

5. Cho, Y.-B., Cho, Y.-H., and Kim, S.-H., “Mining changes in customer buying behavior for collaborative recommendations,” Expert Systems with Applications, 28(2), pp. 359-369, 2005.

6. Han, J., Dong, G., and Yin, Y., “Efficient mining of partial periodic patterns in time series database,” Proceeding of International Conference on Data Engineering, pp. 106-115, 1999.

8. Hung, C.-C., and Peng, W.-C., “A regression-based approach for mining user movement patterns from random sample data,” Data and Knowledge Engineering, 70(1), pp. 1-20, 2011.