Application of Diverse Techniques for Zero-Day Management: A Review Approach

The increasing sophistication and frequency of zero-day attacks pose significant challenges to traditional cybersecurity defences, necessitating innovative approaches for timely detection and mitigation. This study presents a comprehensive systematic literature review on the application of deception security techniques, machine learning algorithms to enhance zero-day attack management. By analysing recent advances in deception-based intrusion control techniques, honeypot and adaptive camouflage, alongside the application of machine learning models in improving threat detection accuracy and response adaptability. Key observations reveal that deception techniques offer early threat detection, behaviour-based analysis, reduced false positives, and effective attack diversion. These attributes make deception a scalable and proactive approach to enhancing cybersecurity defences against zero-day threats. Data of UGRansome was collected which contain classes of zero-day, and then Random forest was selected and trained to generate behavioural analytical model. The model performance recorded 99% accuracy and was validated through comparism with other literatures. The zero-day attack detection model generated was recommended as decision based for deception model for improved zero-day management.