從序列式資料庫挖掘多標籤時間序列樣式

Discovering Multi-label Temporal Patterns in Sequence Database

王昱程(Yu-Cheng Wang)

指導教授 ： 陳彥良

國立中央大學/管理學院/資訊管理學系/碩士(2008年)

摘要

序列樣式探勘是在知識發現和資料挖掘的領域中很重要的一樣技術之一。過去已有很多學者針對序列樣式探勘提出了很多延伸的方法在各種日常應用領域中。以往的研究集中於點式事件和區間式事件或混合事件（包括點式和區間式事件）。然而，在很多日常應用中，事件可能有很多的狀態，不只是發生在某一個時間點或在某一段時間。本研究提出了一個一般化的表達方式來表達時間事件。我們將這些事件視為多標籤的事件，用來代表事件的不同狀態，並提出了MLTPM演算法來從資料庫中挖掘多標籤的時間序列樣式。由於MLTPM是過去方法的一般化，因此它也可以用來處理點式事件，區間式事件和混合事件。實驗結果顯示，MLTPM演算法的效能是可以接受的，且可以找出許多過去方法所忽略的樣式。

關鍵字

區間式事件序列 ； 點式事件序列 ； 多標籤時間序列樣式 ； 時間區間樣式 ； 序列樣式

並列摘要

Sequence pattern mining is one of the most important techniques in knowledge discovery and data mining domain. There were many researches extended the problem of sequential pattern mining in various daily applications. Previous research focused on the point-based event and interval-based event or hybrid event (including point-based and interval-based event). However, in many real life applications, events may have many statuses, not just happens at a certain time point or over a period of time. In this work, we proposed a generalized representation of temporal events. We treated these events as multi-label events which have many statuses, and introduced an algorithm called MLTPM to discover multi-label temporal patterns from temporal database. Since MLTPM is a generalized model of previous methods, it can also deal with point-based events, interval-based events, and hybrid events. The experimental result showed that the MLTPM’s performance is acceptable and can discover interesting patterns.

並列關鍵字

interval-based event sequence ； point-based event sequence ； temporal patterns ； sequential patterns ； multi-label temporal patterns

參考文獻

[3] R. Agrawal, and R. Srikant, “Mining sequential patterns,” Eleventh International Conference on Data Engineering, pp. 3-14, 1995.

[5] J. Allen, “Maintaining knowledge about temporal intervals,” Communications of ACM, vol. 26, no. 11, pp. 832-843, 1983.

[8] Y. L. Chen, M. C. Chiang, and M. T. Kao, “Discovering time-interval sequential patterns in sequence databases,” Expert Systems with Applications, vol. 25, pp. 343-354, 2003.

[9] Y.-L. Chen, and T. C. K. Huang, “Discovering fuzzy time-interval sequential patterns in sequence databases,” IEEE Trans on Systems, Man, Cybernetics- Part B, vol. 35, no. 5, pp. 959-972, 2005.

[10] R. Cooley, B. Mobasher, and J. Srivastava, “Data Preparation for Mining World Wide Web Browsing Patterns,” Knowledge and Information Systems, vol. 1, no. 1, pp. 5-32, 1999.