挖掘频繁波动的Web访问模式算法研究  

Algorithm research on mining frequent fluctuating Web access pattern

作　　者：朱志国[1,2] 邓贵仕[1] 

机构地区：[1]大连理工大学系统工程研究所,辽宁大连116024 [2]东北财经大学津桥商学院,辽宁大连116023

出　　处：《大连理工大学学报》2009年第2期282-287,共6页Journal of Dalian University of Technology

基　　金：国家自然科学基金资助项目(70671016)

摘　　要：考虑到Web访问数据的动态特性,给出了一个从Web访问日志历史演变中挖掘频繁波动的Web访问模式的方法.首先采用无序树结构表示用户历史访问页面序列集合,然后给出了频繁波动Web访问模式的详细定义以及挖掘算法描述,最后,根据数据集中访问序列的大小和数量变化对于算法扩展性和性能的影响进行了实验.结果表明,该算法具备良好扩展性的同时,能够比较高效地提取出频繁波动的Web访问模式.Considering the dynamic feature of Web access data, a method for mining frequent fluctuating Web access pattern （FF-WAP） from historical change of Web users＇ access sequence （WAS） data is presented. Firstly, unordered tree structure is adopted to represent historical WAS sets and then the detailed definition and mining algorithm description of FF-WAP are presented. Finally, according to the number of WAS and the average size of each WAS in data sets, experiments to analyze scalability and performance of the algorithm are conducted. The results show that the algorithm has better scalability and can extract frequent fluctuating Web access patterns efficiently.

关 键 词：数据挖掘 WEB使用挖掘 WEB访问模式 动态数据挖掘 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]