ASPG Menu
search

American Scientific Publishing Group

verified Journal

Journal of Cybersecurity and Information Management

ISSN
Online: 2690-6775 Print: 2769-7851
Frequency

Continuous publication

Publication Model

Open access · Articles freely available online · $500 APC applies after acceptance

Journal of Cybersecurity and Information Management
Full Length Article

Volume 18 • Issue 2 • PP: 01–10 • 2026

An Integrated Framework for Automated Cyber Threat Actor Attribution: LLM-Based Entity Extraction over STIX 2.1 Knowledge Representations with Bayesian Scoring

Aykhan Huseynli 1*
1Department of Cybersecurity, Azerbaijan Technical University, Azerbaijan
* Corresponding Author.
verified

Open Access & Copyright

© 2026 The Author(s). Published by ASPG. This article is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).

Received: May 04, 2026 Revised: August 07, 2026 Accepted: November 01, 2026

Abstract

Cyber threat attribution has become an essential component of national and organizational cyber defense, enabling analysts to link attacks to known threat actors and inform proportionate responses. However, existing attribution practices remain largely non-reproducible, dependent on implicit analyst judgment, and unable to quantify uncertainty in a principled way. The absence of a common intelligence schema further limits cross-organizational comparison and automation. Through a systematic review of attribution methodologies, this study identifies reproducibility, structured representation, and probabilistic calibration as the central unresolved challenges. To address these gaps, this paper proposes an integrated end-to-end framework combining large language model (LLM)-based named entity recognition, STIX 2.1 knowledge representation, and Bayesian probabilistic scoring. The framework implements a unique combination of Campaign Frequency scoring, Inverse Document Frequency-weighted Jaccard similarity, and feature-class-specific Laplace smoothing that correctly distinguishes durable behavioral evidence from perishable technical indicators. The effectiveness of the proposed pipeline is demonstrated using a manually annotated corpus of 25 real-world CTI reports across 15 entity types. The Bayesian attribution engine, evaluated against known threat actors, produces realistic, analyst-interpretable probability distributions.

Keywords

Cyber Threat Intelligence Named Entity Recognition Large Language Models STIX 2.1 Threat Actor Attribution Bayesian Inference

References

[1] T. Rid and B. Buchanan. Attributing cyber attacks. Journal of Strategic Studies, 38(1–2):4–37, 2015.

[2] T. Pahi and F. Skopik. Cyber attribution 2.0: Capture the false flag. In Proceedings of the 18th European Conference on Cyber Warfare and Security (ECCWS 2019), pages 338–345, 2019.

[3] N. Rani, B. Saha, and S. K. Shukla. A comprehensive survey of automated advanced persistent threat attribu-tion: Taxonomy, methods, challenges and open research problems. Journal of Information Security and Applica-tions, 92:104076, 2025.

[4] H. Alturkistani, A. G. Jaafar, and S. Chuprat. Automat-ing cyber threat intelligence workflows with LLMs: Pro-cessing, analyzing, and defending in one model. In 2025 3rd International Conference on Cyber Resilience (ICCR), pages 1–6, Dubai, United Arab Emirates, 2025.

[5] K. Guru, R. J. Moss, and M. J. Kochenderfer. On tech-nique identification and threat-actor attribution using LLMs and embedding models. In 2025 International Conference on Cybersecurity and AI-Based Systems (Cyber-AI), pages 332–339, Varna, Bulgaria, 2025.

[6] V. Mavroeidis and S. Bromander. Cyber threat intel-ligence model: An evaluation of taxonomies, sharing standards, and ontologies within cyber threat intelli-gence. In 2017 European Intelligence and Security Informatics Conference (EISIC), pages 91–98, Athens, 2017.

[7] F. Skopik and T. Pahi. Under false flag: Using technical artifacts for cyber attack attribution. Cybersecurity, 3(1):8, 2020.

[8] A. Saha, J. Mattei, J. Blasco, L. Cavallaro, D. Votipka, and M. Lindorfer. Expert insights into advanced persis-tent threats: Analysis, attribution, and challenges. In 34th USENIX Security Symposium (USENIX Security 25), pages 2185–2204, Seattle, WA, 2025.

[9] D. Blänsdorf. Threat attribution using generative AI: A generative AI-based tool to support the threat attribu-tion process. Master’s thesis, Chalmers University of Technology and University of Gothenburg, Gothenburg, Sweden, 2024.

[10] N. Prasad, A. Diro, M. Warren, and M. Fernando. A survey of cyber threat attribution: Challenges, tech-niques, and future directions. Computers & Security, 157:104606, 2025.

[11] P. Deka, S. Rajapaksha, R. Rani, A. Almutairi, and E. Karafili. AttackER: Towards enhancing cyber-attack attribution with a named entity recognition dataset. arXiv:2408.05149, 2024.

[12] S. Caltagirone, A. Pendergast, and C. Betz. The dia-mond model of intrusion analysis, 2013.

[13] P. N. Bahrami, A. Dehghantanha, T. Dargahi, R. M. Parizi, K.-K. R. Choo, and H. H. S. Javadi. Cyber kill chain-based taxonomy of advanced persistent threat actors: Analogy of tactics, techniques, and procedures. Journal of Information Processing Systems, 15(4):865–889, 2019.

[14] Q. Li, Z. Yang, Z. Jiang, B. Liu, and Y. Fu. Association analysis of cyber-attack attribution based on threat intel-ligence. In Proceedings of the 2nd Joint International Information Technology, Mechanical and Electronic En-gineering Conference (JIMEC 2017), volume 62 of Ad-vances in Computer Science Research, pages 222–230. Atlantis Press, 2017.

[15] Q. Li, Z.-M. Yang, B.-X. Liu, and Z.-W. Jiang. A reasoning method of cyber-attack attribution based on threat intelligence. International Journal of Computer and Systems Engineering, 10(5):920–924, 2016.

[16] K. Edie, C. Mckee, and A. Duby. Extending threat play-books for cyber threat intelligence: A novel approach for APT attribution. In 2023 11th International Sympo-sium on Digital Forensics and Security (ISDFS), pages 1–6, Chattanooga, TN, USA, 2023.

[17] I. Rosenberg, G. Sicard, and E. David. DeepAPT: Nation-state APT attribution using end-to-end deep neu-ral networks. In Artificial Neural Networks and Ma-chine Learning – ICANN 2017, volume 10614 of Lec-ture Notes in Computer Science, pages 91–99, Alghero, Italy, 2017.

[18] M. Alkaabi and S. O. Olatunji. Modeling cyber-attribution using machine learning techniques. In 2020 30th International Conference on Computer Theory and Applications (ICCTA), pages 10–15, Alexandria, Egypt, 2020.

[19] U. Noor, Z. Anwar, T. Amjad, and K.-K. R. Choo. A machine learning-based fintech cyber threat attribu-tion framework using high-level indicators of compro-mise. Future Generation Computer Systems, 96:227–242, 2019.

[20] E. Böge, M. B. Ertan, H. Alptekin, and O. Çetin. Unveil-ing cyber threat actors: A hybrid deep learning approach for behavior-based attribution. Digital Threats, 6(1):1–20, 2025.

[21] N. Xiao, B. Lang, T. Wang, and Y. Chen. APT-MMF: An advanced persistent threat actor attribution method based on multimodal and multilevel feature fusion. Computers & Security, 144:103960, 2024.

[22] X. Xiang, H. Liu, L. Zeng, H. Zhang, and Z. Gu. IPAttributor: Cyber attacker attribution with threat intelligence-enriched intrusion data. Mathematics, 12(9):1364, 2024.

[23] E. Irshad and A. B. Siddiqui. Cyber threat attribution using unstructured reports in cyber threat intelligence. Egyptian Informatics Journal, 24(1):43–59, 2023.

[24] E. Irshad and A. B. Siddiqui. Context-aware cyber-threat attribution based on hybrid features. ICT Express, 10(3):553–569, 2024.

[25] M. R. Ali Khan and A. Almulhem. Behavioral and propagation-based analysis of APT attacks for effective attack attribution. In 2025 13th International Sympo-sium on Digital Forensics and Security (ISDFS), pages 1–8, Boston, MA, USA, 2025.

[26] K. T. W. Teuwen. A modular approach to automatic cyber threat attribution using opinion pools. In 2023 IEEE International Conference on Big Data (BigData), pages 3089–3098, Sorrento, Italy, 2023.

[27] S. Hwang and T.-S. Kim. An exploratory study on artifacts for cyber attack attribution considering false flag: Using delphi and AHP methods. IEEE Access, 11:74533–74544, 2023.

[28] S. Bromander et al. Investigating sharing of cyber threat intelligence and proposing a new data model for en-abling automation in knowledge representation and ex-change. Digital Threats, 3(1):1–22, 2022.

[29] F. Marchiori, M. Conti, and N. V. Verde. STIXnet: A novel and modular solution for extracting all STIX objects in cti reports. In Proceedings of the 18th In-ternational Conference on Availability, Reliability and Security, pages 1–11, Benevento, Italy, 2023.

[30] Y. Ren, Y. Xiao, Y. Zhou, Z. Zhang, and Z. Tian. CSKG4APT: A cybersecurity knowledge graph for ad-vanced persistent threat organization attribution. IEEE Transactions on Knowledge and Data Engineering, 35(6):5695–5709, 2023.

[31] I. J. King, R. Ramirez, B. Bowman, and H. H. Huang. TRAIL: A knowledge graph-based approach for attribut-ing advanced persistent threats. In 2025 IEEE 41st In-ternational Conference on Data Engineering (ICDE), pages 1207–1220, Hong Kong, Hong Kong, 2025.

[32] P. Evangelatos et al. Named entity recognition in cyber threat intelligence using transformer-based models. In 2021 IEEE International Conference on Cyber Security and Resilience (CSR), pages 348–353, Rhodes, Greece, 2021.

[33] S. Zhou, J. Liu, X. Zhong, and W. Zhao. Named entity recognition using BERT with whole word masking in cybersecurity domain. In 2021 IEEE 6th International Conference on Big Data Analytics (ICBDA), pages 316–320, Xiamen, China, 2021.

[34] S.-S. Chen, R.-H. Hwang, C.-Y. Sun, Y.-D. Lin, and T.- W. Pai. Enhancing cyber threat intelligence with named entity recognition using BERT-CRF. In GLOBECOM 2023: 2023 IEEE Global Communications Conference, pages 7532–7537, Kuala Lumpur, Malaysia, 2023.

[35] P. Ranade, A. Piplai, A. Joshi, and T. Finin. CyBERT: Contextualized embeddings for the cybersecurity do-main. In 2021 IEEE International Conference on Big Data (Big Data), pages 3334–3342, Orlando, FL, USA, 2021.

[36] E. Aghaei, S. Jain, P. Arun, and A. Sambamoorthy. SecureBERT 2.0: Advanced language model for cy-bersecurity intelligence. In 2025 Annual Computer Security Applications Conference Workshops (ACSAC Workshops), pages 382–388, Honolulu, HI, USA, 2025.

[37] Y. Ech-Chammakhy, A. Motii, A. Rabii, O. Azrara, and J. Chbili. CyberNER: A harmonized STIX corpus for cybersecurity named entity recognition. In 2025 IEEE 24th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom), pages 2190–2197, Guiyang, China, 2025.

[38] H.-J. Chan, C.-Y. Hsu, C.-C. Chien, J.-J. Wu, and H.-L. Ku. FeedRef2022: A named entity recognition dataset for extracting indicators of compromise. In 2022 IEEE International Conference on Big Data (Big Data), pages 2578–2584, Osaka, Japan, 2022.

[39] S. Della Penna, R. Natella, V. Orbinato, L. Parracino, and L. Pianese. CTI-HAL: A human-annotated dataset for cyber threat intelligence analysis. In 2025 IEEE Eu-ropean Symposium on Security and Privacy Workshops (EuroS&PW), pages 69–78, Venice, Italy, 2025.

[40] D. Plohmann, M. Clauß, S. Enders, and E. Padilla. Mal-pedia: A collaborative effort to inventorize the malware landscape. The Journal on Cybercrime and Digital Investigations, 3(1):1–19, 2017.

Cite This Article

Choose your preferred format

format_quote
Huseynli, Aykhan . "An Integrated Framework for Automated Cyber Threat Actor Attribution: LLM-Based Entity Extraction over STIX 2.1 Knowledge Representations with Bayesian Scoring." Journal of Cybersecurity and Information Management, vol. 18, no. 2, 2026, pp. 01–10. DOI: https://doi.org/10.54216/JCIM.180201
Huseynli, A. (2026). An Integrated Framework for Automated Cyber Threat Actor Attribution: LLM-Based Entity Extraction over STIX 2.1 Knowledge Representations with Bayesian Scoring. Journal of Cybersecurity and Information Management, Volume 18(Issue 2), 01–10. DOI: https://doi.org/10.54216/JCIM.180201
Huseynli, Aykhan . "An Integrated Framework for Automated Cyber Threat Actor Attribution: LLM-Based Entity Extraction over STIX 2.1 Knowledge Representations with Bayesian Scoring." Journal of Cybersecurity and Information Management Volume 18, no. Issue 2 (2026): 01–10. DOI: https://doi.org/10.54216/JCIM.180201
Huseynli, A. (2026) 'An Integrated Framework for Automated Cyber Threat Actor Attribution: LLM-Based Entity Extraction over STIX 2.1 Knowledge Representations with Bayesian Scoring', Journal of Cybersecurity and Information Management, Volume 18(Issue 2), pp. 01–10. DOI: https://doi.org/10.54216/JCIM.180201
Huseynli A. An Integrated Framework for Automated Cyber Threat Actor Attribution: LLM-Based Entity Extraction over STIX 2.1 Knowledge Representations with Bayesian Scoring. Journal of Cybersecurity and Information Management. 2026;Volume 18(Issue 2):01–10. DOI: https://doi.org/10.54216/JCIM.180201
A. Huseynli, "An Integrated Framework for Automated Cyber Threat Actor Attribution: LLM-Based Entity Extraction over STIX 2.1 Knowledge Representations with Bayesian Scoring," Journal of Cybersecurity and Information Management, vol. Volume 18, no. Issue 2, pp. 01–10, 2026. DOI: https://doi.org/10.54216/JCIM.180201
policy

Publisher's Note

The statements, opinions, and data presented in this article are solely those of the author(s) and do not necessarily represent those of ASPG, the journal, or its editors. ASPG and the editors disclaim responsibility for any harm arising from the use of any ideas, methods, instructions, or products described in this article, to the fullest extent permitted by applicable law.

Digital Archive Ready