摘要
通过对挖掘服务器日志文件的分析和研究,可以对网站的组织结构及其性能进行改进,增加个性化服务,发现潜在的读者群体,建立数字图书馆个性化服务的用户模式。数据预处理关系到Web日志挖掘的质量。数据预处理包括数据清理、识别用户、识别用户会话、格式化,目的是分割服务器日志为多个独一无二的用户的一次访问序列。
Thorough analyzing and studying the log of data mining server, we can alter structure and capability of the websites so as to enhance characterized services, find potential users and construct characterized service mode of digital library. Data preconditioning, which includes sequence of data clearing up, user recognition, users＇ conversation recognition and formation, can affect quality of Web log mining and the purpose of it is to divide server log into exclusive sequence of user visit.
出处
《现代情报》
北大核心
2007年第7期65-67,共3页
Journal of Modern Information
