Comparison of Machine Learning and Deep Learning Models for DoS/DDoS Attack Detection

This bachelor's thesis focuses on the detection and classification of DoS and DDoS attacks using machine learning and deep learning methods. The aim of the thesis is to compare selected models from these two approaches in terms of their ability to recognize malicious network traffic and identify individual types of attacks. The theoretical part defines the basic concepts related to DoS/DDoS attacks, describes their principles, classification, impacts, and current defense techniques. It also introduces the principles of machine learning and deep learning, as well as their potential applications in the field of cybersecurity. The practical part is focused on the analysis and preprocessing of the CICDDoS2019 dataset, followed by the simulation of implementation of a two-stage system for attack detection and classification. Selected models were experimentally evaluated using standard classification metrics. The results show that both machine learning and deep learning methods represent effective tools for identifying network attacks, while the individual approaches differ mainly in terms of accuracy, computational complexity, and practical applicability.