Distributed Denial of Service (DDoS) attacks are a formidable network security vulnerability. They disrupt service availability and cause significant damage to organizations. They should be tackled at all costs. This article proposes a proactive approach to detect and distinguish DDoS attacks from typical network traffic instances using machine learning techniques. These techniques include Artificial Neural Network (ANN) and Random Forest. This paper used comprehensive preprocessing techniques, feature engineering strategies, and model architecture to optimize the detection process. The results of this study demonstrated that both Random Forest and ANN models exhibit excellent performance, with an accuracy of 98% and 99%, respectively, in detecting DDoS attacks. The Random Forest model surpasses the ANN model in terms of accuracy, but both models show excellent recall, precision, and F1 scores for classifying typical and DDoS traffic instances.