计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (3): 138-140.DOI: 10.3778/j.issn.1002-8331.2009.03.041

• 数据库、信号与信息处理 • 上一篇    下一篇

IM-FTS:一种快速增量式频繁访问序列挖掘算法

彭慧丽1,张啸剑2,张亚东2   

  1. 1.河南广播电视大学,郑州 450008
    2.河南财经学院 计算机科学系,郑州 450002
  • 收稿日期:2008-01-04 修回日期:2008-04-30 出版日期:2009-01-21 发布日期:2009-01-21
  • 通讯作者: 彭慧丽

IM-FTS:high-speed incremental algorithm for mining frequent traversal sequences

PENG Hui-li1,ZHANG Xiao-jian2,ZHANG Ya-dong2   

  1. 1.Henan Radio & TV University,Zhengzhou 450008,China
    2.Department of Computer Science,Henan University of Finance & Economics,Zhengzhou 450002,China
  • Received:2008-01-04 Revised:2008-04-30 Online:2009-01-21 Published:2009-01-21
  • Contact: PENG Hui-li

摘要: 由于Web数据增长迅速,先前的频繁序列随着序列库的更新而改变。若重新挖掘频繁序列会增加处理时间和数据存储量。提出一种改进的扩展格结构IE-LATTICE,存储先前的挖掘结果,并在其基础上提出一种基于双向约束的增量挖掘算法IM-FTS,在利用先前结果和约束策略前提下,算法仅从插入和删除序列中发现新的频繁序列。分析和实验表明算法能有效缩减数据处理时间和存储空间。

Abstract: Web data grows quickly in the short time,previous FTS may be changed when the sequence database is updated.Re-finding FTS will consume too much execution time and storage space.In this paper,an improved extended lattice,IE-LATTICE is designed to store the previous mining results.An efficient algorithm based on bidirectional constraint,called IM-FTS is proposed,which utilizes the previous results and constraint strategy to discover the new FTS just from the added and deleted part of the database.The analysis and experiments show that IM-FTS algorithm efficiently reduces the average execution time and storage space for mining FTS.