A Master thesis was discussed in Department of Electrical Engineering / College of Engineering at University of Mosul entitled ” Protecting Data Flows in Software Defined Networks (SDN) via Machine Learning Algorithms ” submitted by (Haneen Rafid Mahmood ), Supervised By Assist. Prof. Dr. Mohammed Younis Thanoun on Monday, Oct. 10, 2022.In this thesis, a programmatically defined network topography was built and the reality of a DDOS attack was simulated on one of the connected computers that was considered as a victim computer, and DDOS attacks were detected using the (support vector machine (SVM) algorithm). and it is one of Machine learning algorithms and techniques were also used to mitigate these attacks, namely, a deep packet inspection for a previously collected data set by researchers in the same field, and also attached two types of controllers used in these networks, namely the pox controller and the RYU controller, and the network performance was compared by measuring the throughput and round-trip time and comparing the performance of the controllers.The thesis concluded that RYU was chosen as the controller for the study, as it is better in terms of throughput and round-trip time in the case of the load attack. Round trip 1218,442 milliseconds when using RYU Controller, While the throughput was about 2107 bits/sec when running the simulation for 5 minutes for normal load, while the round-trip time was 13175 milliseconds when using the pox controller, and a Python code was used to measure the accuracy and the detection rate of the attack, where the accuracy value of the test data was 91.8% in While the detection rate is 100% and the false alarm rate is 0%.