ARCHIVES
VOL. 5, ISSUE 2 (2023)
A cloud-native architecture for ai-powered adaptive, learning systems in higher education: design, implementation, and evaluation
Authors
Anoop Sharma
Abstract
The rapid expansion of digital education in
Indian higher education institutions (HEIs) demands scalable, intelligent, and
adaptive learning platforms. This paper presents the design, implementation,
and large-scale empirical evaluation of a cloud-native architecture for
AI-powered adaptive learning systems (AALS). The proposed framework integrates
Kubernetes-based microservices infrastructure with Deep Knowledge Tracing
(DKT), reinforcement learning-based content sequencing, and transformer-based
intelligent tutoring. Evaluated across three Indian universities with 12,847
students over two semesters, the AALS achieved a statistically significant
improvement in academic performance (Cohen's d = 0.56, p <.001), a 34.7%
increase in student engagement, and a 21.3% reduction in dropout rates compared
to a conventional LMS control condition. Total cost of ownership analysis
demonstrated a 41% cost reduction over equivalent on-premise deployments. The
paper contributes a replicable reference architecture and empirical guidelines
for cloud-native AI adoption in resource-constrained HEIs, directly supporting
India's National Education Policy 2020.
Download
Pages:76-83
How to cite this article:
Anoop Sharma "A cloud-native architecture for ai-powered adaptive, learning systems in higher education: design, implementation, and evaluation". International Journal of Educational Research and Development, Vol 5, Issue 2, 2023, Pages 76-83
Download Author Certificate
Please enter the email address corresponding to this article submission to download your certificate.