Leveraging artificial intelligence for cross-framework cybersecurity control mapping: a hybrid natural language processing approach
Journal of King Saud University Computer and Information Sciences · 2026-11 · 38 (9)
DOI: 10.1007/s44443-026-01427-6 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of King Saud University Computer and Information Sciences, online erschienen am 04.11.2026. Verfasst von Hashem Alaidaros. Die Arbeit nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
As cybersecurity standards and regulations continue to evolve, cross-framework control mapping has become more complex for organizations pursuing unified compliance. Existing mapping approaches are largely manual, static, and limited to specific pairs of standards, resulting in high effort and inconsistent outcomes. Although large language models can capture semantic relationships between controls, fully generative approaches may produce inconsistent outputs and provide limited explainability and auditability for compliance-oriented applications. This paper presents an artificial intelligence-enabled framework for mapping cybersecurity controls across heterogeneous standards while accommodating differences in terminology and granularity. The proposed hybrid natural language processing pipeline formulates control mapping as a bidirectional ranking problem and combines control metadata extraction, dense–sparse semantic retrieval, ontology-based similarity, lexical overlap, and large-language-model-assisted aggregation to identify and rank candidate mappings. To improve transparency and auditability, the framework provides decomposed similarity scores, structured relationship classification, and natural-language justifications for each proposed mapping. The framework is evaluated using the Saudi Essential Cybersecurity Controls, published by the National Cybersecurity Authority, and ISO/IEC 27001:2022 as a representative case study. Independent expert validation yields top-1 precision of 98.0% for mapping the Saudi Essential Cybersecurity Controls to ISO/IEC 27001 and 98.5% in the reverse direction. Under an equal-budget evaluation, the proposed framework achieves precision of 90.5% in the first direction and 91.9% in the reverse direction, outperforming dense–sparse embedding baselines (77.0% and 78.9%) and lexical baselines (68.7% and 66.7%). These findings demonstrate how national regulatory requirements can be systematically aligned with international standards through explainable and auditable mapping decisions.
Zitieren
APA
Alaidaros, H. (2026). Leveraging artificial intelligence for cross-framework cybersecurity control mapping: a hybrid natural language processing approach. Journal of King Saud University Computer and Information Sciences, 38(9). https://doi.org/10.1007/s44443-026-01427-6
Vancouver
Alaidaros H. Leveraging artificial intelligence for cross-framework cybersecurity control mapping: a hybrid natural language processing approach. Journal of King Saud University Computer and Information Sciences. 2026;38(9). doi:10.1007/s44443-026-01427-6
BibTeX
@article{Alaidaros2026,
author = {Alaidaros, Hashem},
title = {Leveraging artificial intelligence for cross-framework cybersecurity control mapping: a hybrid natural language processing approach},
journal = {Journal of King Saud University Computer and Information Sciences},
year = {2026},
volume = {38},
number = {9},
doi = {10.1007/s44443-026-01427-6}
}
EUVIMED