摘要
针对序列模式挖掘,提出频繁2序列图(F2SG)来表示数据库中的序列信息,通过扫描一次数据库,将与挖掘任务相关的信息映射到F2SG中,并在此基础上提出一种新的序列模式发现算法——GBSP。GBSP算法充分利用F2SG中表示的项目之间的次序关系进行频繁序列挖掘,提高了其生成效率。理论分析与实验表明,该算法较传统的序列模式发现算法在时间和空间性能上具有优越性。
To the question of sequential pattern mining, this paper introduced frequent 2-sequence graph （F2SG） to express the sequence information related to the mining task by scanning the transaction database only once. Based on F2SG, proposed a new sequential pattern mining algorithm GBSP. The graph representation of database could fully utilize the property of item ordering in the mining process, thus improving the generation efficiency of frequent sequences. Theory analysis and experiment results show that it has better performance in time and space properties than traditional sequential pattern mining algorithms.
出处
《计算机应用研究》
CSCD
北大核心
2008年第4期1003-1005,共3页
Application Research of Computers
基金
国家自然科学基金资助项目(60573174)
安徽省自然科学基金资助项目(050420207)
关键词
数据挖掘
序列模式
图
data mining
sequential pattern
graph
