VESPA (Vehicular Event Sharing with a mobile P2P Architecture) is a system designed for vehicles to share information in vehicular ad-hoc networks (VANETs). The originality of VESPA is to support any type of event (e.g., available parking spaces, accidents, emergency braking, information relative to the coordination of vehicles in emergency situations, etc.). The basic functions of VESPA are both disseminating the events to the potentially interested vehicles and evaluating their relevance once received (for instance to determine whether the driver should be warned or not). In this paper, we concentrate on knowledge extraction in VESPA. In particular, we focus on how to exploit data exchanged between vehicles to produce knowledge to be used later on by drivers. Existing systems only use the data exchanged to produce a warning for the drivers when it is needed. Then, the data is considered obsolete and is deleted. On the contrary, we propose here to aggregate the data once it becomes obsolete. Our objective is to produce additional knowledge to be used by drivers when no relevant data has been communicated by neighbouring vehicles. For example, it becomes so possible to dynamically detect potentially dangerous road segments or to determine the areas where the probability to find an available parking space is high