ADVANCING CYBER THREAT INTELLIGENCE WITH MACHINE LEARNING AND OSINT
DOI: https://doi.org/10.65725/JCISE/2/2/002
JOURNAL OF COMPUTATIONAL INTELLIGENCE SCIENCE AND ENGINEERING (JCISE)
ISSN: 3107-8168
Volume 2 Issue 2, Apr-Jun 2026

Abstract: The growing complexity and tempo of cyber-attacks have outpaced signature-based defenses. Ad-vanced persistent threats, zero-days, and polymorphic malware routinely evade perimeter tools, which creates demand for adaptive, intelligence-driven security. This review proposes a synergistic framework that fuses Machine Learning with Open-Source Intelligence to enable real-time detec-tion and response. We analyze supervised, unsupervised, and deep learning methods for high-volume telemetry, and show how fusing internal analytics with external OSINT improves precision and speeds incident handling. Sector case studies in IIoT, finance, and healthcare illustrate opera-tional value, while challenges in adversarial robustness, data quality, privacy, and model opacity are examined. We outline research directions including Federated Learning for privacy-preserving collaboration, blockchain for trusted threat-intelligence exchange, and Explainable AI for analyst trust. We argue that combining ML with OSINT, augmented by these technologies, is essential to build resilient, transparent, and collaborative cyber defense.

Authors: Spoorthi B S,  Namana D C, Rohan K and Amrutha H P

Keywords: Machine Learning, Block chain, Intrusion detection, Industrial Internet of Things, Cybersecurity frameworks.