Investigation on naturally inspired meta heuristic optimization techniques for web usage mining

Abstract

Web usage mining is data mining technique to investigate the web user behavior with user navigations in web access logs A web log file comprises user patterns and user behavioral patterns for significant identification of similar web users Based on the web user behaviors large quantity of web data is extracted from different web sites Web usage mining carried out with data preprocessing discovery of web patterns and clustering similar web pages Here web usage data s are collected from web browsing history to identify the information depending on their frequent interest Then clustering process is utilized for grouping similar web patterns where data in each cluster allocates a higher similarity Additionally optimization techniques were used in web usage mining for identifying the frequent web usage patterns During the process of clustering similar web patterns accuracy is most essential for better performance of web access patterns Recently many research works have been introduced for enhancing clustering accuracy on web usage patterns However clustering of similar web users still poses the demanding issues in web log files From existing method, in order to overcome clustering issues various web usage mining techniques were developed Existing agglomerative clustering technique was considered to recognize users with similar interest and to establish frequent website visitor newline