Data preprocess in Web log mining
Changsheng Zhou · 2005
Web log mining is one of the most important means of data supporting for Web site management and strategy regulation.The original Web log data are always rough,uncomplete and unstructured,then the quality of preprocessing outcome may greatly affect the quality of mining.Through discussion on the main steps of data preprocessing in Web Usage Mining(WUM),methods of improvement are proposed and the ways of analyzing the information of pictures contained in Web pages are researched.