Artificial Intelligence-Based Approaches for Enhanced Web Personalization: Transforming User Experience and Adaptive Web Interactions

Abstract

Web usage mining is essential for getting user behavior out of Weblogs and customizing business Websites. This work aims to discover these hidden weblog rules by implementing the Apriori Prefix Tree (PT) algorithm on the PSNBC data set. The most important task is to predict the next page based on the user's web activity. The performance of the generated rules was measured with the help of fundamental factors including lift, confidence and support. The findings show that confidence does not vary across the pages but that lift and support correlate highly with the page's significance. Most visited web pages such as News, Front Page, On-air News, Sports and BBS received higher traffic than commonly visited pages like Travel, PSN-News and PSN-Sports. From these findings, it becomes evident that there is potential in web usage mining in consequent log analysis from servers in yielding insightful knowledge for analysis of user conduct and generating personalization content.