Recommender systems have been strongly researched within the last decade. With the emergence and popularization of social networks a new fi eld has been opened for social recommendations. Introducing new concepts such as trust and considering the network topology are some of the new strategies that recommender systems have to take into account in order to adapt their techniques to these new scenarios. In this thesis a simple model for recommendations on twitter is developed to apply some of the known techniques and explore how well the state of the art does in a real scenario. The thesis can serve as a basis for further social recommender system research.