通用Web日志挖掘系统(CWLMS)设计实现  

Designing and Implementation of Common Web Log Mining System

作　　者：刘春梅[1] 

机构地区：[1]华北科技学院,河北三河065201

出　　处：《防灾技术高等专科学校学报》2004年第2期48-52,47,共6页Journal of College of Disaster Prevention Techniques

摘　　要：本文介绍了通用Web日志挖掘系统(CWLMS)的设计与实现。将网站的结构和日志文件格式定义引入Web日志挖掘预处理过程以提高预处理的通用性和准确性;在模式挖掘方面,集成了目前有效的最大向前路径挖掘算法和频繁遍历路径挖掘算法,并且将孤立点分析方法引入日志挖掘中。This paper focuses on the designing and implementation of common web log mining system with three outstanding features. Firstly, the web structure and the definition of log file format are introduced into web log mining preprocessing, which raises the accuracy and efficiency of preprocessing. Secondly, efficient mining algorithm of maximum forward path and efficient mining algorithm of frequent traversal path are integrated in the mining period. Finally, outlier analysis is introduced into the mining system.

关 键 词：数据挖掘 WEB日志挖掘 预处理 孤立点分析 

分 类 号：TP393.4[自动化与计算机技术—计算机应用技术]