Cyber-attacks are increasingly being used and their impact is becoming very dangerous, especially in all digital industries where components are connected via the Internet for communication. These attacks involve multiple connected online devices that are aimed at overwhelming the target server with fake traffic. This scenario, in most cases, results in a destructive process for the target server. DDoS assaults do not trying to break through your security circumference; otherwise, they aim at making your website unavailable for users. Also, this type of attack can be used to cover up other malevolent activities. This research introduces the use of a machine learning algorithm to detect DDoS. In addition, we used multiple SDN datasets from different data sources. The aim of that is to integrate different data and use cases as much as we can to enrich our model with different scenarios. We utilized an MLP machine learning algorithm in our model to detect the malicious behavior of DDoS attacks. Likewise, an input layer of 19 neurons was used in our model to be equal to the number of features in the dataset, and great accuracy results were achieved.