Predictive Modeling for Cyber Threat Intelligence

EasyChair Preprint 14525

Abstract

Cybersecurity is increasingly becoming a critical area of focus due to the rise in sophisticated cyber threats. Traditional threat detection methods often fail to address the dynamic and evolving nature of modern cyber-attacks. To mitigate these threats, predictive modeling has emerged as a powerful approach within Cyber Threat Intelligence (CTI). This abstract outlines the potential of predictive models to anticipate and prevent cyber-attacks by analyzing patterns and trends in vast datasets, utilizing machine learning (ML) algorithms and artificial intelligence (AI).

Predictive modeling in CTI leverages historical cyber incident data, threat actor behaviors, and network traffic patterns to generate actionable insights. By identifying correlations and emerging threats, these models provide real-time assessments, enabling organizations to bolster their defenses proactively. The key methods include supervised learning, which helps classify known threats, and unsupervised learning, which detects anomalies or zero-day exploits. Techniques like regression analysis, clustering, and neural networks form the backbone of these predictive systems, empowering cybersecurity analysts to stay ahead of threat actors.

Keyphrases: Cybersecurity, Threat Intelligence, predictive modeling

@booklet{EasyChair:14525,
  author    = {Favour Olaoye and Kaledio Potter},
  title     = {Predictive Modeling for Cyber Threat Intelligence},
  howpublished = {EasyChair Preprint 14525},
  year      = {EasyChair, 2024}}