User-modelled web browsing

Abstract

In this project we will be looking at ways to analyse an individual's internet usage on a daily basis upon which a user model will be generated. The use of the internet has been increasing exponentially these last few years. As a consequence users are being inundated with too much information, losing sight of the useful information among the rest. A tool embedded inside the browser will identify the pages visited by the user and determine their field and category. Upon continuous use, this system will model the user's preferred categories and feed this information to the recommendation system. This system will be able to provide the user with other pages which they may be interested in with respect to their user model and their preferred categories. This will allow the user to explore new different, yet related pages relevant to their liking. It would allow the user to be presented with relevant pages without the information overload of irrelevant results.