How to calculate trust between social network users?

Trust among users in a social network plays an important role when trying to build a social recommender system. To be able to calculate who-trusts-whom and implement that knowledge in the social recommender, it is crucial to examine in detail all possible relations between users in the social network and to evaluate those relations properly. We propose a model to transform a single user's social graph from a binary structure to a structure with the concrete weights between the nodes in a graph. This paper first addresses a question of how can a general model for calculating trust be implemented using a specific social network as a medium (i.e., Facebook). Afterwards, the proposed model is verified through a Facebook application named “Closest Friends”, which calculates the user's closest friends on the Facebook social network. Finally, our approach has been evaluated with the experiment among the 150 Facebook users who tested “Closest Friends” application.