AI-DRIVEN INTRUSION DETECTION SYSTEMS FOR NEXT-GENERATION CAMPUS NETWORKS: A LAYERED INTELLIGENCE FRAMEWORK
Keywords:
Intrusion Detection, Machine Learning, Anomaly Detection, University Network Security, AI Security, Behavioural Analysis, Knowledge Graph, Next-Generation EngineeringAbstract
As university campuses become more digitally dependent, their network environments grow increasingly complex and difficult to secure using traditional rule-based intrusion detection methods. Higher education institutions face a rapidly escalating threat landscape: a 2025 global analysis recorded a 63% year-over-year surge in verified university cyberattacks, with 425 significant incidents across 67 countries [11], while 91% of UK higher education institutions reported at least one confirmed breach in the prior twelve months [12]. Static signature databases, threshold-based alerting, and manual log analysis cannot keep pace with these threats. This paper proposes an AI-Driven Intrusion Detection System (AI-IDS) with three intelligence layers: a real-time anomaly detection engine using supervised machine learning, a behavioural profiling module using unsupervised clustering, and a threat correlation layer powered by a knowledge graph. The architecture is evaluated against five realistic university threat scenarios, with integration considerations for SIEM platforms, concrete hardware specifications, and GDPR-compliant privacy governance.
