Securing Networks: Leveraging Machine Learning for Enhanced DDoS Detection
Som Chauhan, Prajval Byahatti, Sunil Kumar Patel · 2024
The growing complexity of Distributed Denial of Service (DDoS) attacks poses a serious threat to network security, making the development of effective detection and mitigation techniques essential. This study tackles this issue by introducing a novel approach that utilizes machine learning (ML) to improve the identification of DDoS attacks using realworld datasets. By leveraging the CSE-CIC-IDS2018 AWS dataset, we develop a strong classifier to examine network traffic patterns that suggest potential DDoS activity. Our approach encompasses comprehensive steps, including data collection, preprocessing, and the deployment of sophisticated ML algorithms. The outcomes reveal impressive performance, with the model achieving a $99.98 \%$ accuracy rate and nearperfect precision and F1 scores. These results highlight the strength of neural network-based models, especially when implemented using advanced frameworks such as Keras. This research ultimately bolsters network defenses against DDoS attacks and emphasizes the necessity of ongoing innovation in detection techniques to protect digital resources and maintain uninterrupted online services in our increasingly digitalized environment.