Online services are the basis of the current, digital world. Operations using distributed denial of service (DDoS) remain a significant threat to the availability of online services. DDoS attacks have become increasingly dangerous with the rapid development of computers and communications technology. At present, there has not yet been a detection method with satisfactory detection accuracy due to the diversity of DDoS attack modes and the large volume of traffic involved in DDoS attacks. Therefore, this paper proposes a machine-learning-based method for detecting DDoS attacks. The proposed technique created a confusion matrix to identify the model performance after applying the machine learning models. The first classification uses the KNN classifier technique for both Precision (PR) and Recall (RE) then the Precision (PR) and Recall are both classified using the DNN classifier technique in the second classification (RE). According to the experiment results, the suggested machine learning-based DDoS attack detection system has a high detection rate for the current DDoS attack which is most prevalent and the results are implemented by using python software.