Knowledge Discovery from Web Data for Web Personalization

Abstract

Because of the large and rapid increase in web data size and number of users, web users now face the problems of overloading and drowning information. As a result, Recovery of internet-based data and web applications, providing web users with more accurate information becomes a critical issue. In this study, by analyzing web data features, we aim to improve the performance of web information retrieval and web presentation through web data mining processes that discover the knowledge (intrinsic relationships) between web data expressed as textual, linkage or usability information. We concentrate on discovering web usage patterns through web usage mining, and then using the discovered usage knowledge along with profile information to provide web users with more personalized web content. Personalization is an engaging service for website visitors, based on their characteristics and deliberate behaviors to facilitate conversion and long-term commitment expectations. The purpose of this work is to extract the knowledge from web data and use this knowledge to create a web personalization system that allows users to access the content of their need from the website without specifically specifying it. The knowledge could be the navigational actions of the user as exposed by web access log analysis, as well as the characteristics and preferences of the user reflected by user profiles. Such knowledge is further analyzed to improve system performance, retention of users and/or modification of the site. This paper provides a comprehensive survey of the different approaches suggested by Web Personalization researchers and list out some of the issues that need to be tackled soon