AI-BASED ZERO-DAY ATTACK PREDICTION TECHNIQUES IN NETWORK, WEB APPLICATION, AND SYSTEMS
Research Hub · 01.01.2026 · 07 (02) · 584-601
DOI: 10.63665/rh.v7i2.87 ↗Kurz gefasst
Fachartikel in der Zeitschrift Research Hub, online erschienen am 01.01.2026. Verfasst von Ruchika Parate und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Zero-day attacks pose a significant threat to modern digital infrastructures due to their ability to exploit previously unknown vulnerabilities and evade traditional signaturebased security mechanisms. The increasing complexity and interconnectedness of networked systems, web applications, and host environments have further intensified the need for proactive and intelligent defense strategies. In recent years, Artificial Intelligence (AI) has emerged as a promising approach for predicting zero-day attacks by learning behavioral patterns and identifying anomalies indicative of malicious activity. This paper presents a comprehensive study of AI-based zero-day attack prediction techniques across network, web application, and system-level environments. A structured taxonomy is introduced to categorize existing approaches based on learning paradigms, model architectures, and deployment layers. The study further provides a comparative analysis of machine learning, deep learning, and hybrid techniques, highlighting their effectiveness, limitations, and domain-specific challenges. In addition, a unified conceptual architecture is proposed to integrate cross-layer intelligence and enhance early prediction capabilities through behavioral correlation. The analysis reveals that while AI-driven models demonstrate strong potential in identifying unknown attack behaviors, challenges related to data scarcity, false positives, model interpretability, and scalability continue to affect real-world deployment. The findings of this research offer valuable insights for researchers and practitioners and contribute toward the development of adaptive, reliable, and proactive cybersecurity solutions for mitigating zero-day threats in complex digital environments.
Zitieren
APA
Parate, R., Dandhare, A., Saklani, N., & Lokhande, S. (2026). AI-BASED ZERO-DAY ATTACK PREDICTION TECHNIQUES IN NETWORK, WEB APPLICATION, AND SYSTEMS. Research Hub, 07(02), 584-601. https://doi.org/10.63665/rh.v7i2.87
Vancouver
Parate R, Dandhare A, Saklani N, Lokhande S. AI-BASED ZERO-DAY ATTACK PREDICTION TECHNIQUES IN NETWORK, WEB APPLICATION, AND SYSTEMS. Research Hub. 2026;07(02):584-601. doi:10.63665/rh.v7i2.87
BibTeX
@article{Parate2026,
author = {Parate, Ruchika and Dandhare, Aarti and Saklani, Nikita and Lokhande, Shubham},
title = {AI-BASED ZERO-DAY ATTACK PREDICTION TECHNIQUES IN NETWORK, WEB APPLICATION, AND SYSTEMS},
journal = {Research Hub},
year = {2026},
volume = {07},
number = {02},
pages = {584--601},
doi = {10.63665/rh.v7i2.87}
}
EUVIMED