Intelligent bookmark organiser and recommender

Abstract

This dissertation aims to investigate the possibility of adding unsupervised features to a user’s web browsing experience through bookmark organisation and recommendation. Organisation with and without user defined labels is investigated using techniques such as the Vector Space Model and Latent Semantic Analysis as well as a multitude of clustering algorithms. Diverse methods for bookmark recommendation based on the Vector Space Model are also explored. An artefact in the form of a browser extension, having the aforementioned features of organisation and recommendation, was produced and handed out to test subjects to gather feedback and ultimately gauge the extent to which bookmarks were organised and recommended to users’ satisfaction.