Universal Cyber Attack Prediction System (UCAPS): A Framework for Predictive Cyber Defence Using Threat Intelligence, Knowledge Graphs, Attack Graph Analytics and Artificial Intelligence
Year: 2026
Type: preprint
Abstract: Contemporary cybersecurity technologies support threat detection, incident response, and recovery, although their capabilities and emphasis vary across tools and operational environments. Security Information and Event Management (SIEM), Endpoint Detection and Response (EDR), Extended Detection and Response (XDR), threat intelligence platforms, and security orchestration solutions provide important capabilities for security monitoring and response. Forecasting future attacks before exploitation remains a challenging research problem. Modern cyber threats, including ransomware campaigns, Advanced Persistent Threats (APTs), cloud-centric attacks, supply-chain compromises, and AI-assisted adversarial operations, motivate further investigation of predictive defensive capabilities. This paper proposes the Universal Cyber Attack Prediction System (UCAPS), a conceptual framework for investigating how cyber-attacks might be forecast before exploitation occurs. The framework integrates Threat Intelligence, Vulnerability Intelligence, Behaviour Analytics, Cyber Knowledge Graphs, Attack Graph Analytics, Artificial Intelligence, and Cyber Digital Twins within a proposed predictive cyber defence architecture. UCAPS considers attack probability estimation, attack-path forecasting, exploitability prediction, target-asset prediction, and simulation-assisted defensive planning. Mathematical formulations, research hypotheses, evaluation criteria, and a multi-phase research roadmap are presented. The framework is proposed as a foundation for research into a transition from predominantly reactive incident management toward predictive and intelligence-driven cyber defence. This paper presents a proposal; it does not report an implemented or experimentally validated UCAPS system.
Author: Angajala, Srinivasa Rao
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7235249470
