User Identification, Classification and Recommendation in Web Usage Mining - An Approach for Personalized Web Mining
P. Priyanga, N C Naveen · 2015
In recent years, Web Analytics (WA) is turning out to be an emerging research topic due to the extensive advancements in the techniques that aid in accessing the web contents, which millions of people have shared on the web. The information that has connection to the theme being searched may not be recognized always, if the personalization system operates in accordance with the usage-dependent outcomes alone. In this research work a new method is introduced for Personalized Web Search system, wherein, the users are enabled to have access to the relevant web pages as per their choice from the URL list. The first stage of this research deals with Semantic Web Personalization, which provides a merging between the content semantics as well as the usage data that are stated as ontology terms. This system supports the computation of the navigational patterns that are semantically improvised, so that constructive recommendations can be successfully engendered. It can be perceived that no other systems excluding the semantic web personalization system described here is employed in nonsemantic web sites. The second stage of the work is to assist in augmenting the quality of the recommendations depending on the structure lying beneath the website. Finally, the testing is achieved through the utilization of a prolonged database link. The analysis of variation that exists among the different classes of parameters is made later, when the privacy is formulated using the memory usage and the period of execution.