Abstract. Vehicular ad hoc networks (VANETs) consist of a set of nodes com- posed by devices present in vehicles and along the roads, that communicate with each other. VANET applications offer more entertainment, comfort, safety and other benefits to drivers and passengers. However, like any wireless network, they present security challenges for vehicular applications. Currently, one of the major security challenges in VANETs is to handle the distributed denial-of- service (DDoS) attacks that affect the data and network availability. The main purpose of this work is to detect DDoS attacks in VANETs using machine learn- ing. The proposed solution uses a VANET hybrid architecture in fog computing and using Support Vector Machine (SVM).