Empowering Cybersecurity: An Adaptive Approach With Hybrid Machine Learning For Anomaly Detection

Abstract

The fast development in the use of computer networks raises concerns about network availability, integrity, and confidentiality. This requires network managers to use various types of intrusion detection systems (IDS) to monitor network traffic for unauthorized and malicious activity. In this research, a hybrid machine learning-based framework is introduced for anomaly detection in the system. The suggested hybrid machine learning model, consisting of C4.5, a convolutional neural network (CNN), and a random forest (RF), was applied to the Bot-IoT dataset. The proposed hybrid intrusion detection framework achieved 99.8% accuracy, 96.4% precision, 100% recall, and an F1 score of 98.1% in the classification of malicious activities. This research suggests a more reliable and comprehensive approach to managing the ever-changing landscape of cyber threats by demonstrating the exceptional performance of the proposed framework.