Web Usage Behavior Pattern Analysis and Recommender System

Abstract

Almost anything we do online leaves traces of our activities on the Internet. These footprints offer an opportunity to study various aspects of human behavior. Aim of this research is to analyze web usage behavior patterns to promote self-awareness that helps bring positive changes in individual’s performance. Time spent at browsing particular web page is a key metric to estimate the website usage. In this thesis, we describe the framework that first collects and processes the data including quantitative data such as time spent, number of visits and qualitative data i.e. web site category. Second, we describe the web usage behavior modeling to extract valuable and interesting temporal and categorical patterns regarding user interests, peak browsing time of day, most visited websites, related websites groups according to website categories, frequent tab switching, and session’s statistics i.e. number of sessions per day, number of tabs per session etc. To discover the valuable behavior patterns from the individual’s browsing data, different web usage mining techniques have been used including statistical analysis, associative rule mining and clustering. Finally, we demonstrate interactive visualizations for the analysis and monitoring of web browsing behavior patterns with the goal of providing the individual with detailed understanding of behavior, provide recommendations and present the social comparison framework to promote competition and motivation among individuals that can bring out the ambition and push one that’s good for person’s personal and professional growth. The experimental results demonstrate interesting correlations of web categories usage with different times of the day, project deadlines and user productivity. Questionnaire is drawn to evaluate the proposed system ‘BBA’. According to survey results, 90% of users found ‘BBA’ very effective that made them conscious and aware of their web usage behavior.