摘要
在日志数据的预处理中,确定合适的挖掘粒度是一项重要任务。本文介绍了一种日志数据预处理模型,在一般预处理模型基础上添加了页面视图识别环节,从而使日志数据有了更精确的挖掘粒度,挖掘结果有更强的语义。
Finding a proper mining granularity is a crucial task, which should be finished in log data preproeessing, This paper puts forward an improved preprocessing model of Web usage data. An additional step, page view identification is appended to the common model, then the improved model is realized through experiment, and finally, from the sample result, the granularity is detailed and is more meaningful than before.
出处
《计算机与现代化》
2007年第4期62-63,104,共3页
Computer and Modernization
基金
上海高校选拔培养优秀青年教师科研专项基金资助项目(沪教委[2005]80号)
关键词
WEB挖掘
预处理
日志挖掘
页面视图识别
Web mining
preprocessing
usage data mining
page view identification