Explainable AI in Intrusion Detection for Resource-Constrained Organizations: A SHAP-LLM Framework with Adversarial Robustness Evaluation

Forthcoming, under peer review

Chinemelum Umealajekwu

Forthcoming

Combines SHAP explanations with a language model to make intrusion detection decisions understandable for teams without dedicated security analysts, and evaluates how the system holds up against adversarial attacks.

Osholake, Umealajekwu, Edohen, Majekodunmi, Evans-Anoruo

IRE Journals · Vol. 8 Iss. 6, pp. 1102–1115 · December 31, 2024

Proposes the CLOSO (Cognitive Load Optimized Security Operations) framework for how AI and human analysts can work together in a security operations center to reduce analyst workload and cognitive fatigue.

Paper ID: IRE 1709110 · ISSN 2456-8880

View publication

Cited in ACM Transactions on Internet Technology.