基于三元互关联后继树的Web日志挖掘  

WEB LOG MINING BASED ON TERNARY INTER-RELATED SUFFIX TREE

作　　者：张忠平[1] 高一博[1] 

机构地区：[1]燕山大学信息科学与工程学院,河北秦皇岛066004

出　　处：《计算机应用与软件》2011年第10期50-54,共5页Computer Applications and Software

基　　金：国家自然科学基金(60773100);河北省教育厅科研计划项目(2006143)

摘　　要：针对二元的互关联后继树模型进行改进,构造三元互关联后继树,并结合Web日志的特点,构造Web事务集的互关联后继树和增量更新模型,设计基于三元互关联后继树的频繁路径挖掘算法,挖掘Web日志中的频繁路径。通过实验证明了基于三元互关联后继树和改进后的模型的出色插入查询性能。In this paper we constructed a ternary inter-related suffix tree based on the improvement of binary inter-related suffix tree model,and constructed the inter-related suffix tree of web transactions and its incremental refreshing model in combination with the characteristics of Web logs,we also devised a frequent paths mining algorithm based on the ternary inter-related suffix tree to mine the frequent paths in Web logs.Through the experiments we demonstrated the superb performance of insertion query of the improved model which is based on the ternary inter-related suffix tree.

关 键 词：WEB日志挖掘 WEB事务 互关联后继树 频繁路径 

分 类 号：TP311.132.4[自动化与计算机技术—计算机软件与理论]