Mashup FOAF for Video Recommendation LightWeight Prototype

Abstract

There are more and more xml document, web services, feeds and so on and so forth cheap, network accessible resources to use. As one of the most widely used semantic web project, FOAF (Friend of a Friend) pays more and more attention to FOAF semantic features to analyze users’ interest and to recommend to FOAF users recent years. This essay focuses on applying FOAF to a latest online television programs recommendation system for a particular user. In this paper television programs come from various online video web sites that a user has registered in are watched at different time. The article describes the approach to such services based on HMM (Hidden Markov Model) and FOAF project. In order to protect the user’s privacy when providing services, this system is designed as a local-service desktop model. We conduct experiments to illustrate users’ high degree of satisfaction to our techniques.