一种重构网站结构的Web日志挖掘数据预处理方法  

Data Pre-processing in Web Log Data Mining Based on Reconstructed Web Structure

作　　者：袁健[1] 金鑫[1] 

机构地区：[1]上海理工大学,上海200093

出　　处：《小型微型计算机系统》2011年第7期1427-1430,共4页Journal of Chinese Computer Systems

基　　金：上海市重点学科建设项目(T0502)资助

摘　　要：在Web日志挖掘的过程中,数据预处理是整个Web日志挖掘过程的基础,其直接影响了日志挖掘的质量和结果.由于目前大多数网页都采用框架模式,而传统的预处理技术并没有针对frame页面进行过滤,即使过滤,也会导致页面结构的混乱,从而不能够为路径补充提供正确的信息.基于此,本文提出一种基于重构网站结构的Web日志挖掘数据预处理方法以及基于它的路径补充方法.In the process of web log mining, data preprocessing is the basis for the entire mining process, which directly affects the quality and results log mining. Most Web pages are based on the framework model, the traditional pre-processing technology has not filtered the frame pages, even if doing ,the filter would lead to confusion in the page structure, so as not to be able to provide the correct information for the complementary path. Based on this, this paper proposes a log data preprocessing based on the reconstructed site web methods and a complementary path＇s method based on it.

关 键 词：网站重构 数据预处理 WEB 151志挖掘 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]