Web user navigation patterns discovery from WWW server log files

Continued growth of user number and size of shared content on Web sites cause the necessity of automatic adjusting content to users' needs. In the literature of Web Mining, such actions are referred to personalization and recommendation which led to improve the visibility of presented content. To perform adequacy actions which correspond to the expected users' needs we can utilize web server log files. Mining such data with accurate constraints can lead to the discovery of web user navigation patterns. Such knowledge is used by personalization and recommendation systems (PRS) due to performed actions against user behavior during a visit on the web portal. In these paper we present the system framework for mining web user navigation patterns in order to knowledge management. We focus on constraints which are critical factors to evaluate the effectiveness of the implemented algorithm. On the other hand, these constraints can be perceived as knowledge validation criteria due to its adequacy. Thus only adequate knowledge can be added to existing in PRS knowledge base.