K M, Zubair and Akhtaruzzaman, Khan and Tanvir Rahman, Akash (2022) BUILDING TRUST IN AUTONOMOUS CYBER DECISION INFRASTRUCTURE THROUGH EXPLAINABLE AI. International Journal of Economy and Innovation, 29. pp. 405-428. ISSN 2545-0573

Text 405-428+BUILDING+TRUST+IN+AUTONOMOUS+CYBER+DECISION.pdf Download (1MB)

Abstract

The high rate of the development of digital technologies has changed cyber security from a manual, rule-based practice to a complex, data-driven field that needs intelligent automation. The study article, which has the title of Building Trust in Autonomous Cyber Decision Infrastructure through Explainable AI (XAI), investigates the potential of explain ability to promote transparency, reliability and trust in human operators to AI-based cyber security systems. Conventional black-box AI models are accurate, but they are frequently uninterpretable, which creates distrust in the cyber security community, which bases its opinions on understandable arguments to confirm automated warnings. This study incorporates Explainable AI (SHAP, SHapley Additive Explanations, and LIME, Local Interpretable Model-Agnostic Explanations) into the machine learning models to interpret network anomalies, protocol behavior, and intrusion patterns with the help of the Network Intrusion Detection Dataset. These findings indicate that XAI is not only more interpretable but also it narrows the divide between AI decision-making and human monitoring, and leads to operational trust and responsibility. Among the essential conclusions, it is possible to note that model transparency has a direct impact on operator trust, which allows implementing timely, effective, and auditable responses to cyber threats. Moreover, explain ability helps in determining key characteristics such as connection duration, failed logins, and the use of protocols, which contribute to the detection of anomalies, thereby enhancing the accuracy of analysts and human perception. The paper concludes that using XAI in Autonomous Cyber Decision Infrastructures (ACDI) is not only capable of making cyber security defenses predictive, but also intelligible and ethically sound. This study has added to the current development of transparent, adaptive, and intelligent cyber security systems that can effectively address the dynamic threat environment through increased trust, accountability and human-AI interactions.

Item Type:	Article
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:	Postgraduate > Master's of Islamic Education
Depositing User:	Journal Editor
Date Deposited:	18 Oct 2025 09:21
Last Modified:	18 Oct 2025 09:21
URI:	http://eprints.umsida.ac.id/id/eprint/16432

Actions (login required)

View Item