Denial of service (DDoS) attacks, popularly known as Distributed Denial of Service, represent an enormous threat to not only companies but also the individuals who depend on them. This kind of assault denies the network’s intended user access to its services culminating in a time and money waste. Even though DDoS attacks are not new to traditional networks, they have emerged as a serious security threat to SDN-enabled networks. To study this dataset which is related to DDoS-affected SDN, different machine learning techniques are used including Logistic Regression with Random Forest (LR-RF), K-Nearest Neighbors with Support Vector Machine (KNN-SVM), Multilayer Perceptron with Gated Recurrent Unit (MLP-GRU), Iterative Dichotomiser 3 with Classification and Regression Tree (ID3-CART), and Stochastic Gradient Descent with Adaptive Moment Estimation (SGD-ADAM). Also included is a comparison and analysis of harmful data types based on protocol such as UDP flood attacks, TCP SYN assaults, and ICMP attacks. In addition, ID3-CART algorithm achieved accuracy rates as high as 99.95% in detecting malicious traffic based on experimental results. Area Under the Curve of Receiver Operator Characteristic (AUC-ROC) curves are used to evaluate and validate the models in this research. To mitigate and identify DDoS in network management in SDN systems, the best approaches for identifying DDoS attacks are therefore recommended through the methodology provided in this study.