Network administration has become more challenging due to the increasing variety of Internet applications and the ongoing development of network infrastructure with new technologies. The ability to classify network traffic is essential for managing network resources in accordance with security and quality of service standards. Nevertheless, the strict scalability, security, and privacy requirements are not addressed by the conventional traffic classification techniques based on Deep Packet Inspection. Software-Defined Networking’s centralised controller facilitates traffic analysis and gives direct programming capabilities while offering a global picture of the network. It is possible to dynamically modify traffic flows to meet evolving network needs. These qualities, in conjunction with the use of machine learning techniques, have made it feasible to optimise, better manage, and maintain networks, as well as integrate intelligence into them.