In this thesis, we propose a light-weight collaborative filtering method for supporting the implementation of pervasive recommender systems on current and common mobile devices. To achieve this, we propose a distributed collaborative filtering-based recommender system that does not explicitly model the user in terms of his or her preferences, yet still delivers results according to them. The system we use to demonstrate this upon is a restaurant guide, one which is 'grown' by its users, by relying upon them to be the suppliers of a comprehensive list of interesting restaurants and their experiences with them. We describe how this information is added, and how it is then used via our recommender system to assist other users in identifying an appropriate choice of restaurant quickly and accurately using current, ubiquitous, infrastructure elements.
Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:587516 |
Date | January 2011 |
Creators | Sklenar, Lukas |
Publisher | University of Kent |
Source Sets | Ethos UK |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Page generated in 0.0059 seconds