Serendipitous Recommendations in Social Networks
PD Dr. Georg Groh
Recommender systems use a variety of methods to recommend items that can be characterized as serendipitous. Serendipitous recommendations can be assumed to be of special value for the user. Thus it is often the goal to maximize serendipty. Social Networking as a paradigm of Social Computing allows for determining and integrating social context, e.g. in the form of social networks into services. The research question investigated in this thesis is in how far social context, especially social networks, can be used in recommender systems for Social Networking platforms in order to maximize serendipity. The thesis will first review the state of the art in terms of serendipity for recommender systems in social networks. Using a design science methodology, a solution concept will be developed.