EUVIMEDEUVIMED entdecken
Crossref

A Hybrid Graph Neural Network and Large Language Model Framework for Insider Threat Detection via Behavioral Graph and Semantic Profiling

Md Abu Sufian Mozumder, Mohammad Musa Mia, Rumana Akther Nipa, Asaduzzaman Anik, klachur Rahman Bhuiyan, Sharmin Akter, Mashaeikh Zaman Md. Eftakhar Choudhury

International Journal of Computer Science & Information System · 11.05.2026 · 11 (05) · 22-32

DOI: 10.55640/ijcsis/volume11issue05-03 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computer Science & Information System, online erschienen am 11.05.2026. Verfasst von Md Abu Sufian Mozumder und 6 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Insider threats remain one of the most critical and elusive challenges in enterprise cybersecurity due to their ability to exploit legitimate access while evading traditional detection mechanisms. In this study, a hybrid framework integrating Graph Neural Networks and Large Language Models is proposed to enhance insider threat detection through the fusion of behavioral graph modeling and semantic profiling. Using the CERT Insider Threat Dataset and the UEBA Dataset on Kaggle, the model captures both relational dependencies among users, devices, and resources, and contextual insights from unstructured textual data such as logs and communications. The experimental results demonstrate that the proposed hybrid model significantly outperforms traditional machine learning, sequence-based, and single-modality deep learning approaches, achieving an accuracy of 0.96, an F1-score of 0.92, and a ROC-AUC of 0.95. These improvements are primarily driven by the model’s ability to jointly learn structural anomalies and semantic deviations, enabling more accurate detection of multi-stage and stealthy insider attacks. Furthermore, the integration of explainable language-based outputs enhances interpretability and operational usability in enterprise security environments. The findings highlight the effectiveness of multi-modal learning in advancing insider threat detection and provide a scalable, practical solution for deployment in large-scale enterprise systems.

Zitieren

APA

Mozumder, M. A. S., Mia, M. M., Nipa, R. A., Anik, A., Bhuiyan, K. R., Akter, S., & Choudhury, M. Z. M. E. (2026). A Hybrid Graph Neural Network and Large Language Model Framework for Insider Threat Detection via Behavioral Graph and Semantic Profiling. International Journal of Computer Science & Information System, 11(05), 22-32. https://doi.org/10.55640/ijcsis/volume11issue05-03

Vancouver

Mozumder MAS, Mia MM, Nipa RA, Anik A, Bhuiyan KR, Akter S, et al. A Hybrid Graph Neural Network and Large Language Model Framework for Insider Threat Detection via Behavioral Graph and Semantic Profiling. International Journal of Computer Science & Information System. 2026;11(05):22-32. doi:10.55640/ijcsis/volume11issue05-03

BibTeX
@article{Mozumder2026,
  author = {Mozumder, Md Abu Sufian and Mia, Mohammad Musa and Nipa, Rumana Akther and Anik, Asaduzzaman and Bhuiyan, klachur Rahman and Akter, Sharmin and Choudhury, Mashaeikh Zaman Md. Eftakhar},
  title = {A Hybrid Graph Neural Network and Large Language Model Framework for Insider Threat Detection via Behavioral Graph and Semantic Profiling},
  journal = {International Journal of Computer Science & Information System},
  year = {2026},
  volume = {11},
  number = {05},
  pages = {22--32},
  doi = {10.55640/ijcsis/volume11issue05-03}
}
Kostenlos starten →