Different Aspects of Web Log Mining

The expansion of the World Wide Web has resulted in a large amount of data that is now freely available for user access. The data have to be managed and organized in such a way that the user can access them efficiently. For this reason the application of data mining techniques on the Web is now the focus of an increasing number of researchers. One key issue is the investigation of user navigational behavior from different aspects. For this reason different types of data mining techniques can be applied on the log file collected on the servers. In this paper three of the most important approaches are introduced for web log mining. All the three methods are based on the frequent pattern mining approach. The three types of patterns that can be used for obtain useful information about the navigational behavior of the users are page set, page sequence and page graph mining.