ENHANCING ACCESS CONTROL AND ANOMALY DETECTION IN DIGITAL EDUCATION SYSTEMS THROUGH ZERO TRUST AND MACHINE LEARNING INTEGRATION

Omar J. Alkhatib; Zoha Farooq; Nadeem Arif; Asma Javaid

Authors

Omar J. Alkhatib
Zoha Farooq
Nadeem Arif
Asma Javaid

Abstract

Purpose

Recent developments in digital education platforms have increased exposure to cybersecurity issues, making the traditional perimeter-based security frameworks inadequate. The study will explain how effectively Zero Trust Architecture and machine learning-based anomaly detection can be used together to enhance access control, security, and adoption readiness in digital education systems.

Methodology

The research design was quantitative and cross-sectional, where a structured questionnaire was conducted in digital education settings among stakeholders. A total of 300 respondents who were students, instructors, IT administrators, and management personnel in different educational institutions were sampled to gather data. The survey assessed perceptions of Zero Trust implementation, machine learning anomaly detection, security and privacy issues, and readiness to adopt on a five-point Likert scale. Reliability, descriptive statistics, Pearson correlation, multiple regression analysis and one-way ANOVA were used to analyze data.

Findings

As the results depict, the measurement tool is internal-consistent with a combined Cronbachs alpha of 0.93, which proves the soundness of the assessment of the entire constructs. The implementation of Zero Trust Architecture (M = 4.02) and Machine Learning-based anomaly detection (M = 3.95) were highly agreed by the respondents, which indicates that they have a positive view about the effectiveness of these in digital education systems. Correlation analysis showed that there is a significant positive relationship between Zero Trust implementation and adoption readiness (r = 0.72), and Zero Trust and ML anomaly detection (r = 0.68). The outcomes of multiple regression indicate that the strongest predictor of the adoption and readiness is Zero Trust implementation (β=0.42), then ML anomaly detection (β=0.31), and security and privacy perceptions play a minor role (β=0.14). Role-based analysis provides a higher level of readiness to adopt among management and instructors than among students, which is why the level of engagement will differ between stakeholder groups.

Implications

The results indicate that Zero Trust can be combined with Machine Learning into a versatile and dynamic security system of digital education systems. The research offers practical information on how educational institutions can become more resilient to cybersecurity and at the same time ensure its usability and user confidence.

Originality/Value

The study will present empirical evidence on user perceptions and adoption willingness of intelligent, Zero Trust-based security framework in educational settings. The research is based on the central user-centered approach to improving the security of online learning through a combination of architectural security measures and information-based anomaly detection.

Keywords: Zero Trust Architecture, Machine Learning, Anomaly Detection, Digital Education Systems, Access Control, Cybersecurity, Adoption Readiness