Personalized TV with Recommendations : Integrating Social Networks

Sammanfattning: This master’s thesis concerns how to recommend multimedia content which a user might view – with some media player. It describes how a computer application (called a recommendation engine) can generate better recommendations for users based on using information available from social networks and the media selections made by others. This thesis gives an introduction to the area of “recommendations”, recommendation engines, and social networks. An overview of existing recommendation techniques suggests potential solutions to the problem of what recommendations to make to a given user. The thesis presents how social networks can be used to further enhance the users’ experience and describes the work that has been done to realize this recommendation system. An evaluation of the implemented solution is given. The thesis concludes with a summary of how recommendation engines and social network technologies can be used and suggests some future work. This thesis is of current interest since there is a tremendous quantity of content which is being offered in stores and via services on the web. A recommendation system makes it easier for users to find content which they find appropriate. Since social network communities are growing rapidly there has been an interest to use this information to get recommendations from friends. The results from the evaluation of the prototype recommendation system show how social networks which utilize trust systems might affect the recommendation which is given.