A distributed denial-of-service (DDoS) attack is a deliberate attempt to stop a target system, service, or network from operating normally by flooding it with an avalanche of unauthorized traffic or requests. In a DDoS assault, numerous hijacked computers or devices—often forming a botnet—are organized to simultaneously send the target an excessive amount of traffic or requests, overwhelming it and making it unable to react to valid user requests. DDoS attacks, such as volumetric attacks, TCP/IP protocol attacks, application layer attacks, and reflective/amplified attacks, can take many forms and target distinct weaknesses in the target infrastructure. DDoS attack prevention might be difficult, but there are different ways of detecting a DDoS attack. This paper focuses on finding and reviewing different kinds of ML-based DDoS detection techniques suggested in different papers and journals and comparing these methods and finding the best alternate solution.