Early Prediction of Denial-of-Service Attacks on Campus Area Network Using Ensemble Learning

A Denial of Service (DoS) attack aims to prevent a network or online service from functioning properly by overloading it with traffic or by exploiting security flaws. Machine learning algorithms are used for recognizing DoS attack on the Campus Area Network in order to circumvent these denial-of-service attacks. By analyzing data trends, machine learning algorithms can determine the contents of each packet. The approach presented in this research study, employs ensemble learning trained machines with more sophisticated artificial intelligence to prevent denial of service attack. In this research paper, an ensemble learning system is created using machine learning models to conclude the optimal solution for attack prediction in campus area network. This mechanism is designed to stop attacks at early basis. Due to its usage of the best features from several models, ensemble recognition is more reliable and accurate. The research demonstrates how well an ensemble learning approach offers early indicators of impending denial-of-service attack on a for campus area network. The results prove that ensemble learning with different maximum depth levels leads to better results in terms of ROC Curve, recall, and accuracy, as well as the F1 score, when detecting whether traffic patterns comprise attack packets or benign packets. The results of this study add to current efforts to improve network security and stability in educational and community settings.