Enormous amounts of data are permanently generated by customers of telecommunication services and their interactions. These data can be used to provide personalized recommendations of mobile services. In this master thesis a comprehensive user model for such a recommender system will be developed. The aim is to identify appropriate features that can be utilized within sophisticated recommendation approaches. These features will be obtained by studying the state-of-the-art in this field as well as by conducting empirical analyses of existing data of an Austrian Telco. Statistical methods and machine learning techniques will be applied. For this master thesis a part-time employment within a project of TU Wien and the mentioned Austrian telecommunications company can be offered.