挖掘闭合多维序列模式的可行方法  

Feasible method for mining closed multidimensional sequential pattern

作　　者：纪兆辉[1] 李存华[1] 

机构地区：[1]淮海工学院计算机工程学院,江苏连云港222005

出　　处：《计算机工程与设计》2009年第22期5065-5067,5074,共4页Computer Engineering and Design

基　　金：2008年度江苏省高校自然科学基础研究基金项目(08KJD520013)

摘　　要：为了对闭合多维序列模式进行挖掘,研究了多维序列模式的基本性质,进而提出了挖掘闭合多维序列模式的新方法。该方法集成了闭合序列模式挖掘方法和闭合项目集模式挖掘方法,通过证明该方法的正确性,指出闭合多维序列模式集合不大于多维序列模式集合,并且能够覆盖所有多维序列模式的结果集。最后分析了该方法所具备的两个明显优点,表明了在闭合多维序列模式挖掘中的可行性。To mine the closed multidimensional sequential pattern, some basic properties of multidimensional sequential pattern are firstly studied. Then, a new method for mining closed multidimensional sequential pattern is proposed. This method is an integration of closed sequential pattern mining and closed itemset pattern mining. By proving the correctness of this method, conclusions are drawn that the set of closed multidimensional sequential patterns is not larger than the set of multidimensional sequential patterns and the set of colsedmultidimensionalsequentialpatternscancovertheresultingsetofmultidimensionalsequentialpatterns. Finally, two advantages of this method are analyzed and the feasibility of this method for mining closed multidimensional sequential pattern is verified. Key words： data mining; closed sequential pattern; closed itemsets pattern; multidimensional pattern; closed multidimensional pattern

关 键 词：数据挖掘 闭合序列模式 闭合项目集模式 多维模式 闭合多维模式 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]