Recommender Systems (RSs) are software tools and techniques providing suggestions for items, such as movies, CDs, or travels, to be of use to a user. In general, a recommendation can be more compelling and useful if the context of the user is known. For instance, in a travel recommender, the season of the travel, or the group composition, or the motivation of the travel are all important contextual factors that, as a traveler normally does, should be taken into account by a system to generate more relevant recommendations. In this paper we show how a context-aware mobile recommender system for places of interest (POIs) selection can generate more effective recommendations than those produced by a non context-aware version, i.e., those normally provided to the city visitors by the local tourist office. Here we mainly focus on the HCI solutions and in particular in the explanation of the recommendations that are perceived by the user as an important element of the graphical interface.