Neural Sentinels: Intelligent Threat Hunting in the Age of Autonomous Attacks
1 Master of Science in Information Technology, Washington University of Science and Technology, Alexandria, Virginia, USA.
² Bachelor of Education, National University, Bangladesh.
³ Master's in Genetics, Osmania University, India.
⁴ Master’s in Commerce, Jagannath University College, Dhaka, Bangladesh.
⁵ Bachelor in Law, Independent University Bangladesh.
⁶ Master's in Information Technology, Washington University of Science and Technology, Virginia, USA.
⁷ Professor of Cybersecurity, Daffodil International University, Dhaka, Bangladesh.
Review Article
World Journal of Advanced Research and Reviews, 2022, 16(03), 1480-1488
Publication history:
Received on 18 November 2022; revised on 25 December 2022; accepted on 28 December 2022
Abstract:
Autonomous cyber-attacks are growing at a fast pace, and such a rapidly changing nature makes rules-based mechanisms invalid in this field. Automation, AI-powered reconnaissance, and adjustable attack vectors that evade static detection systems are more frequently utilized by contemporary threat actors. We introduce a threat-hunting framework called Neural Sentinels that uses supervised machine-learning models to detect malicious activities using behavioral and contextual features. By examining a structured cybersecurity dataset with user risk metrics, device trust scores, failed login attempts, and DNS tunneling indicators, we assess Logistic Regression, random forest, gradient boosting, support vector machine (SVM), and K-Nearest Neighbors (KNN). Experimental findings show that SVM performs best, with 94.5% accuracy and 0.989 ROC-AUC, limited in comparison to the ensemble and linear baselines. The results indicate that intelligent detection systems based on behavior are a key component in improving the resilience under autonomous attacks.
Keywords:
Threat hunting; Machine learning; Autonomous attacks; Intrusion detection; Cybersecurity analytics; Neural Sentinels
Full text article in PDF:
Copyright information:
Copyright © 2022 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0