An effective web page reorganization through heap tree and farthest first clustering approach

Web usage mining is used to extract interesting and useful information from a web server log file. The log file is automatically stored in web server. It contains details about the web user activity on the website. Web users' interest changes over time, which is not stable. Hence, the static website will get outdated soon. So Website needs to be modified with minimal changes to meet user requirements. The objective of this research is to retrieve required web pages with minimal search cost in a website. Preprocessing is done to get the required data. An algorithm Max Heap with Farthest First Clustering Approach (HFCA) is proposed in which Farthest first clustering algorithm is used to cluster frequently accessed web pages and then the web pages are reorganized using the max heap tree. Reorganization is based on the frequently accessed web pages. Experimental results shows that max heap tree along with the farthest first clustering gives better performance and minimal search cost between the web pages. It is suitable for the dynamic website that needs a change periodically.