Intrusion Detection on Self Organizing Network using PCA and Random Forest

There are several hazards to internet security as a result of the development of network technology. One cannot stress how crucial intrusion detection systems (IDS) are to network and computer security. Addressing the challenge of class imbalance in intrusion detection data, this study explores the development of an efficient Intrusion Detection System (IDS). Using the KDDCUP'99 dataset, we propose an IDS based on the Random Forest classification algorithm enhanced with Principal Component Analysis (PCA). The system's performance was evaluated using accuracy, precision, recall, F-score, and Confusion Matrix analysis. The diverse precision levels inherent in different IDS methods underscore the need for a reliable and effective solution.