Abstract

Intrusion Detection Systems (IDS) are critical components of modern cybersecurity infrastructure designed to monitor detect and prevent malicious activities in networked environments. The application of machine learning algorithms particularly Support Vector Machines (SVM) has significantly enhanced the detection accuracy and efficiency of IDS by enabling automated classification of normal and anomalous network traffic. This research focuses on a Polynomial Kernel Support Vector Machine (SVM) trained on data collected from intrusion detection systems to classify network behavior and detect potential security breaches. The study examines the mathematical foundations of polynomial kernel SVMs their applicability to IDS datasets feature selection techniques hyperparameter optimization and performance evaluation metrics such as accuracy precision recall and F1-score. Furthermore the research analyzes challenges associated with high-dimensional data class imbalance computational complexity