时序关系下的闭合序列模式挖掘算法-北京邮电大学学报.PDF

2013年8月 北 京 邮 电 大 学 学 报 Aug.2013 第36卷 第4期 Journal of Beijing University of Posts and Telecommunications Vol.36 No.4 摇 摇 文章编号:1007鄄5321(2013)04鄄0019鄄04 DOI:10.13190/ jbupt.201304.16.fuy 时序关系下的闭合序列模式挖掘算法 1 1 1 2 付摇 宇 ,摇 于艳华 ,摇 宋美娜 ,摇 战晓苏 (1.北京邮电大学PCNCAD 中心,北京 100876;2.军事科学院军事运筹分析研究所,北京 100876) 摘要:序列挖掘算法产生冗余序列,造成其运行时间过长. 对此,提出了一种新的闭合序列挖掘算法———时序关系 下的闭合序列模式挖掘算法. 依据闭合序列模式的性质,通过比较频繁序列与每个1鄄项频繁序列之间的时序关系, 推断频繁序列模式是否可扩展. 基于IBM公司的合成数据,将其与闭合序列模式挖掘算法进行比较,实验结果表 明,这种新的闭合序列挖掘算法可以有效降低运行时间且不易受到属性值个数的影响. 关摇 键摇 词:时序关系;闭合序列模式;数据挖掘 中图分类号:TN929郾53摇 摇 摇 摇 文献标志码:A A Closed Sequential Pattern Mining Algorithm in Time Order 1 1 1 2 FU Yu ,摇 YU Yan鄄hua ,摇 SONG Mei鄄na ,摇 ZHAN Xiao鄄su (1.PCNCAD Center,Beijing University of Posts and Telecommunications,Beijing 100876,China; 2.Institute of Military Operations Research and Analysis,Academy of Military Science,Beijing 100876,China) Abstract:Sincethereexistredundantsequentialpatternsinresults,suchminingrunsforalongtime. To combat this drawback,a new algorithm,called closed sequential patterns mining algorithm in time order (CloTSP) ,is proposed. Based on the nature of closed sequential patterns,CloTSP canjudge whether a frequent sequential pattern is extended by comparing its time order with each frequent 1鄄item subse鄄 quence. Experiments on the synthetic data produced by International business machines corporation se鄄 quential patterngenerator show that CloTSPcanshortenrun鄄timesignificantly comparedtoclosedsequen鄄 tial pattern mining (CloSpan). Furthermore,it is also shown that the run鄄time of