Summary: | 碩士 === 國立中央大學 === 資訊管理研究所 === 92 === Sequential pattern mining is of great importance in many applications including computational biology study, consumer behavior analysis, system performance analysis, etc. Recently, an extension of sequential patterns, called time-interval sequential patterns, is proposed by Chen, Jiang, and Ko, which not only reveals the order of items but also the time intervals between successive items. For example: having bought a laser printer, a customer returns to buy a scanner in three months and then a CD burner in six months. Although time-interval sequential patterns are useful in predicting when the customer would take the next step, it can not determine when the next k steps will be taken. Hence, we present two efficient algorithms, MI-Apriori and MI-PrefixSpan to solve this problem. The experimental results show that the MI-PrefixSpan algorithm is faster than the MI-Apriori algorithm but the MI-Apriori algorithm has a better scalability.
|