Risk Mitigation Approach to Cyber Threat using AI-Driven Models for the Evolving Threat Landscape.
Publication Date: 17/01/2025
Author(s): Jesufemi Olanrewaju, Matthias Oluloni Togunde, Oyebola Akande .
Volume/Issue: Volume 8 , Issue 1 (2025)
Abstract:
This systematic review examines the effectiveness of AI-driven models in mitigating evolving cyber threats, using the PRISMA framework to analyze studies published between 2019 and 2024. The review focuses on machine learning techniques, including supervised, unsupervised, and deep learning. Findings show that deep learning excels in detecting complex threats like Advance Persistent Threats (APTs) and zero-day vulnerabilities, while supervised learning is effective for known threats but struggles with new attack types. Unsupervised learning adapts well to dynamic environments but has higher false positive rates. The review proposes a multi-layered framework combining AI models with traditional security measures for enhanced threat detection and response. A hybrid approach is recommended as the most effective strategy, though challenges like data quality and algorithmic bias must be addressed for optimal implementation.
Keywords:
Advance Persistent Threats, zero-day, PRISMA framework, multi-layered framework, AI-driven
