The improvements in vehicle transmission technologies and machine learning, smart transportation networks, have become an important part of the growth of modern cities. This study looks at how Vehicular Ad-hoc Networks (VANETs) and the Internet of Vehicles (IoV) work together, focusing on how they can make transportation systems safer, more reliable, and more efficient. VANETs are a specific type of Mobile Ad-hoc Networks (MANETs) that let cars talk to each other and to infrastructure. This lets people share information and make decisions in real time. The IoV builds on this idea by connecting vehicle networks to larger IoT systems. This makes the transportation environment smarter and more linked. Machine learning (ML) has had a big effect on VANETs and IoV by providing more advanced ways to look at data, make predictions, and improve performance. ML algorithms improve many parts of smart transportation networks, such as managing traffic flow, preventing accidents, and finding the best routes. The review carefully looks at the current state of study on VANETs and IoV, with a focus on how ML methods are used. ML integration for better network performance, better vehicle-to-everything (V2X) connection, and better route methods are some of the most important areas of research. The review also discuss about the problems and restrictions of using ML in these networks, like worries about data privacy, limited computer resources, and the need for strong algorithms that can work in settings that are always changing for vehicles. This review puts together the results of several recent studies to give a full picture of how ML can be used to deal with the difficulties of smart transportation networks.