STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal Of Network security Original Research

AI-Driven Cyber Threat Detection and Response Framework for Higher Educational Institutions (HEIs)

  1. Irfan Belim Faculty of Computer Application, Noble University
  2. Bhargav Rajyagor Faculty of Computer Application, Noble University

Abstract

Higher educational institutions (HEIs) are increasingly exposed to complex cyber threats due to their open-access digital ecosystems, diverse user populations, and extensive research infrastructures. Conventional signature-based intrusion detection systems often fail to detect zero-day and evolving attacks, creating an urgent need for intelligent and adaptive security mechanisms. This research introduces an AI-based framework for detecting and responding to cyber threats, specifically designed for HEIs. The framework employs an unsupervised machine learning model—Isolation Forest—trained exclusively on benign network traffic to identify anomalous behavior without prior attack signatures. Using the CSE-CIC-IDS2018 dataset, the model was evaluated within a dynamic monitoring window of 50,000 network flows. Experimental results identified 2,115 anomalous flows, corresponding to a 4.23% anomaly rate. Unlike static evaluation models, the proposed framework incorporates a dynamic risk scoring mechanism that adapts to real-time institutional traffic patterns by comparing current anomaly rates against historical baselines. The observed anomaly deviation remained within acceptable operational thresholds, indicating a low institutional threat level. The findings demonstrate that AI-driven anomaly detection can effectively support proactive cybersecurity strategies in HEIs. The proposed framework offers both practical and conceptual improvements by combining machine learning-driven anomaly detection with cyber risk assessment techniques for institutions.

Keywords

References (17)

  1. Almuhanna R, Dardouri S. A deep learning/machine learning approach for anomaly based network intrusion detection. Front Artif Intell. 2025;8:1625891. doi:10.3389/frai.2025.1625891. PMID:40995028.
  2. Neto ECP, Iqbal S, Buffett S, Sultana M, Taylor A. Deep learning for intrusion detection in emerging technologies: a comprehensive survey and new perspectives. Artif Intell Rev. 2025;58:340. doi:10.1007/s10462-025-11346-z.
  3. Alabdulatif A. A novel ensemble of deep learning approach for cybersecurity intrusion detection with explainable artificial intelligence. Appl Sci (Basel). 2025;15:7984. doi:10.3390/app15147984.
  4. Wu Y, Zou B, Cao Y. Current status and challenges and future trends of deep learning-based intrusion detection models. J Imaging. 2024;10:254. doi:10.3390/jimaging10100254. PMID:39452417.
  5. Sajid M, Malik KR, Almogren A, Malik TS, Khan AH, Tanveer J, et al. Enhancing intrusion detection: a hybrid machine and deep learning approach. J Cloud Comput. 2024;13:123. doi:10.1186/s13677-024-00685-x.
  6. Tian J. Integrating artificial intelligence into the cybersecurity curriculum in higher education: a systematic literature review. Educ Sci (Basel). 2025;15:1540. doi:10.3390/educsci15111540.
  7. Adewusi MA. A design-science, conceptual framework paper proposing a zero-trust security architecture and implementation roadmap for AI-enabled cybersecurity in university-industry digital ecosystems within the Industry 5.0 era [Preprint]. 2025. doi:10.2139/ssrn.5837705.
  8. Afolalu O, Tsoeu MS. Cybersecurity in higher education institutions: a systematic review of emerging trends, challenges and solutions. Future Internet. 2025;17(12):575. doi:10.3390/fi17120575.
  9. Bwiino K, Mayoka GK, Nkamwesiga L, Nyamadi M, Shirani P. A systematic literature review of information security practices in higher education contexts. IET Inf Secur. 2026;2026(1):1-19. doi:10.1049/ise2/6324508.
  10. Mahuta GA, Abubakar H. Artificial intelligence in higher education in Nigeria: challenges and way forward. J Contemp Res Educ Adm Manag. 2025;2:62-71.
  11. Ibrahim N, Rajalakshmi NR. Examining the influence of advanced persistent threats on higher education institutions and investigating appropriate cybersecurity strategies. U.Porto J Eng. 2025;11:96-119. doi:10.24840/2183-6493_0011-002_002671.
  12. Liu FT, Ting KM, Zhou ZH. Isolation forest. In: Proceedings of the Eighth IEEE International Conference on Data Mining; 2008 Dec 15-19; Pisa, Italy. p.413-22. doi:10.1109/ICDM.2008.17.
  13. Sharafaldin I, Gharib A, Lashkari AH, Ghorbani AA. Towards a reliable intrusion detection benchmark dataset. Softw Netw. 2018:177-200.
  14. Wada IU, Izibili GO, Babayemi T, Abdulkareem A, Macaulay OM, Emadoye A. AI-driven cybersecurity in higher education: a systematic review and model evaluation for enhanced threat detection and incident response. World J Adv Res Rev. 2025;25:2233-45.
  15. Singh NK, Panday A, Shrivastava V. AI in cybersecurity: detecting and preventing cyber threats using machine learning. Clareus Sci Sci Eng. 2025;2(8):19-25.
  16. Reddy Y. Artificial intelligence in intrusion detection systems: trends, frameworks, and future directions for cybersecurity. Int J Intell Syst Appl Eng. 2023;12:949-59.
  17. D'Hooge L. CSE-CIC-IDS2018 [dataset]. Kaggle; 2018. Available from: https://www.kaggle.com/datasets/dhoogla/csecicids2018
Support