Home / Current Issue / Paper 1709633
Advanced Threat Detection in Zero Trust Architectures: A Machine Learning Approach
Subject area: Science,Engineering and Technology · Area of research: Machine Learning
Abstract
This study investigates the integration of machine learning algorithms within zero trust security frameworks to enhance threat detection capabilities. Using a dataset of 1.2 million network events collected from three enterprise environments, we evaluate six supervised and unsupervised learning techniques for identifying anomalous behavior patterns that indicate potential security breaches. The research specifically focuses on optimizing the balance between minimizing false positives and maintaining detection sensitivity. Our findings demonstrate that ensemble models combining deep learning with traditional detection methods achieve up to 96.7% accuracy while reducing false positives by 73.4% compared to conventional rule-based systems. This research provides empirical evidence supporting the efficacy of machine learning-augmented zero trust architectures for advanced threat detection in modern enterprise environments.
Keywords
Machine Learning, Zero Trust, Threat Detection, Anomaly Detection, Ensemble Models, Enterprise Security
How to cite this paper
@article{1709633,
author = {Job Adegede},
title = {Advanced Threat Detection in Zero Trust Architectures: A Machine Learning Approach},
journal = {Iconic Research And Engineering Journals},
year = {2025},
volume = {9},
number = {1},
pages = {1456-1469},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1709633.pdf},
abstract = {This study investigates the integration of machine learning algorithms within zero trust security frameworks to enhance threat detection capabilities. Using a dataset of 1.2 million network events collected from three enterprise environments, we evaluate six supervised and unsupervised learning techniques for identifying anomalous behavior patterns that indicate potential security breaches. The research specifically focuses on optimizing the balance between minimizing false positives and maintaining detection sensitivity. Our findings demonstrate that ensemble models combining deep learning with traditional detection methods achieve up to 96.7% accuracy while reducing false positives by 73.4% compared to conventional rule-based systems. This research provides empirical evidence supporting the efficacy of machine learning-augmented zero trust architectures for advanced threat detection in modern enterprise environments.},
keywords = {Machine Learning, Zero Trust, Threat Detection, Anomaly Detection, Ensemble Models, Enterprise Security},
month = {July},
}