An architectural framework for web information retrieval based on user´s navigational pattern

As the popularity of WWW explodes, a massive amount of data is gathered by Web servers in the form of Web access logs which are rich source of information for understanding the Web user´s surfing behavior. Web Usage Mining is an application of data mining algorithms to Web access logs to find trends and regularities in Web users´ traversal pattern. Web page prediction technique is an important research area in web technologies. Mining is useful for web path traversal pattern from web logs. Web traversal pattern mining discovers most of the users´ access patterns from web logs. This information can provide the routing suggestions for web users such that suitable actions can be adopted. However, the web data will grow speedily in the short time, and some of the web data may be out of date. The user behaviors may change when old web data in web logs is replaced by new data. Therefore, we must re-discover the user behavior from the updated web logs but, it is very time-consuming. In this work we present efficient algorithms for web page prediction from large web logs visited by a user. We assign a significant weight to each page based on time spent by user on each page and visiting frequency on each page.