REVOLUTIONIZING DIGITAL TRUST: NETWORK INTRUSION DETECTION SYSTEMS FOR IDENTITY AND SECURITY ASSURANCE IN THE METAVERSE

Digital security has become a major challenge in the metaverse, an interactive virtual space that integrates augmented reality and virtual reality. This study develops a machine learning-based Network Intrusion Detection System (NIDS) to enhance security reliability within the metaverse. K-Means and Apriori algorithms are applied to optimize rules in the Snort IDS, enabling more accurate detection of Distributed Denial of Service (DDoS) and Malware Command and Control (CNC) attacks. The results show that rule optimization using machine learning increases detection accuracy for DDoS attacks from 60% to 75% and for CNC attacks from 35% to 40%. Furthermore, this approach successfully reduces the false positive rate. The implementation of the optimized NIDS provides a significant contribution to securing activities in the metaverse, ensuring a safer and more reliable virtual environment.