<![CDATA[This paper presents a systematic literature review (SLR) on AI-based algorithms for cybersecurity threat detection, aiming to evaluate the effectiveness and performance differences of various artificial intelligence techniques. The purpose of this study is to provide a comprehensive overview of the most effective AI models for detecting cyber threats and to examine their practical applications across various cybersecurity domains, including IoT, critical infrastructure, and cyber-physical systems. The review includes studies published between 2021 and 2025, sourced from prominent academic databases such as MDPI, SpringerLink, and IEEE Xplore.The methodology employed involved the selection of peer-reviewed articles using inclusion and exclusion criteria, followed by thematic analysis of the AI techniques used in the studies. Key themes such as supervised learning, unsupervised learning, deep learning, and hybrid approaches were explored. Performance metrics including accuracy, precision, recall, F1-score, and false positive rates were used to evaluate algorithm effectiveness. The results highlight the comparative performance of different AI models and provide insights into the strengths and weaknesses of each approach, as well as their suitability for specific cybersecurity applications.The findings emphasize the importance of dataset quality, algorithm transparency, and the need for reducing false positives in real-world applications. The review concludes by recommending the continued development of hybrid AI approaches and the need for more transparent, explainable models.]]>
Copyrights © 2025
