In the vehicle domain, a recommender system (RS) helps vehicle items to be displayed and selected replied on the parameters that are the most relevant to specific users. In other words, users can meet with vehicle items suited to their profile, preference, or interest. Therefore, users will not spend much time searching and retrieving what corresponds to their needs and capacities, as well as against the ever-growing volume of information in the application. In this paper, we realize the analysis of the particular characteristics of users and vehicle items in the vehicle recommendation. As a result, we propose our approach for building an RS based on ontologies in the vehicle domain.