User frequent navigation pattern mining model for web log

Conventional Web mining models for web log almost only consider the distance between pages and ignore the structure of a Web site, so they can't do data mining accurately. For a Web site framework, almost Web pages near to the homepage are navigating pages and those far, from homepage are content pages. For analyzing the navigating pages, access times is more dominant, while for analyzing content pages, access time length is more dominant. A new Web mining model for user frequent navigation pattern is issued, which considers the Web site framework adequately. It is proved that the model figures out the shortages of conventional Web mining models for user frequent navigation pattern by experimentations. It can be used to improve E-business Web site framework and efficiency of the Web site system to make users more satisfactory.