Network Intrusion Detection Using Supervised Machine Learning for DDoS Attack Classification

The rapid growth of internet-based services has significantly increased the exposure of network infrastructures to sophisticated cyber threats, particularly Distributed Denial-of-Service (DDoS) attacks that disrupt service availability and compromise organizational operations. Traditional signature-based intrusion detection systems (IDS) struggle to identify evolving and zero-day attacks, necessitating intelligent and adaptive detection mechanisms. This paper proposes a supervised machine learning–based intrusion detection framework for accurate classification of malicious network traffic. Traffic data from the KDD Cup 99 benchmark dataset are preprocessed using feature selection and normalization techniques to improve model efficiency and reduce computational overhead. Multiple classifiers, including C4.5 Decision Tree and Naïve Bayes, are trained and evaluated for attack detection performance. Experimental results demonstrate that the Decision Tree model achieves superior detection accuracy and lower false-positive rates compared to probabilistic approaches.