Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM67819
Volume:
10 |
10 |
Oct 2026
Pages:
1-4
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Department of Computer Application
jaihind institute of managment & research,kuran
Department of Computer Application
jaihind institute of managment & research,kuran
Most colleges and universities today use Learning Management Systems (LMS) like Moodle and Canvas to deliver course materials and conduct online assessments. However, these platforms typically deliver identical content to all students, ignoring differences in their learning speeds, prior knowledge, and individual difficulties. While machine learning can help personalize education through adaptive content recommendations and early dropout prediction, deploying these models in production environments introduces serious practical challenges. These include high server response times during concurrent examinations, train-test feature skews in streaming data, student data privacy concerns under FERPA/GDPR regulations, and opaque 'black-box' predictions that teachers cannot interpret or act upon. In this paper, we propose a scalable, cloud-based intelligent e-learning architecture designed to address these limitations. The proposed system separates concerns into four distinct layers: learner interaction interfaces, a distributed cloud data pipeline using Kafka and object storage, a containerized AI inference engine, and a teacher supervision dashboard with human-in-the-loop controls. Student clickstream events and quiz logs are captured in real time to trace concept mastery using Deep Knowledge Tracing, recommend tailored exercises using contextual multi-armed bandits (LinUCB), and alert educators about students at risk of academic failure using gradient-boosted trees explained via TreeSHAP. We also define a comprehensive evaluation methodology combining normalized learning gains, recommendation accuracy, P99 latency benchmarks, and demographic fairness metrics to support responsible real-world university deployment.
KEYWORDS
In this paper, we presented a cloud-based intelligent e-learning architecture that brings together adaptive content recommendation, early failure prediction, and teacher oversight. Drawing on published research across educational data mining, artificial intelligence, and cloud engineering, our framework addresses common real-world challenges including high server latency, inflexible linear workflows, and opaque black-box predictions. By combining streaming data pipelines, knowledge tracing within the zone of proximal development, and transparent teacher dashboards, the system allows universities to support individual student needs at scale. The mathematical models and multi-dimensional evaluation plan outlined here provide a practical, realistic roadmap for educational institutions looking to modernize their online learning environments.
[1] T. Kabudi, I. Pappas, and D. H. Olsen, "AI-enabled adaptive learning systems: A systematic mapping of the literature," Comput. Educ. Artif. Intell., vol. 2, p. 100017, 2021. [2] C. Gligorea, M. Cioca, R. Oancea, A. T. Gorski, H. Gorski, and P. Tudorache, "Adaptive learning using artificial intelligence in e-learning: A literature review," Educ. Sci., vol. 13, no. 12, p. 1216, 2023. [3] F. Wang, D. Zou, H. Xie, G. Zou, and F. L. Wang, "Artificial intelligence in education: A systematic literature review," Expert Syst. Appl., vol. 255, p. 124467, 2024. [4] M. A. Almaiah, A. Al-Khasawneh, and A. Althunibat, "E-learning-based cloud computing environment: A systematic review, challenges, and opportunities," IEEE Access, vol. 12, pp. 7329–7355, 2024. [5] A. T. Corbett and J. R. Anderson, "Knowledge tracing: Modeling procedural knowledge," User Model. User-Adapt. Interact., vol. 4, no. 4, pp. 253–278, 1994. [6] C. Piech, J. Bassen, J. Huang, S. Ganguli, M. Sahami, L. J. Guibas, and J. Sohl-Dickstein, "Deep knowledge tracing," in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 28, pp. 505–513, 2015. [7] J. Kuzilek, M. Hlosta, and Z. Zdrahal, "Open University Learning Analytics Dataset," Sci. Data, vol. 4, p. 170171, 2017. [8] J. L. Rastrollo-Guerrero, J. A. Gómez-Pulido, and A. Durán-Domínguez, "Analyzing and predicting students' performance by means of machine learning: A review," Appl. Sci., vol. 10, no. 6, p. 2027, 2020. [9] S. M. Lundberg and S.-I. Lee, "A unified approach to interpreting model predictions," in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 30, pp. 4765–4774, 2017. [10] M. Armbrust, A. Ghodsi, R. Xin, and M. Zaharia, "Lakehouse: A new generation of open data platforms," in Proc. Conf. Innovative Data Syst. Res. (CIDR), 2021.
Publication Date
10/12/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
1
Total views since publication
PDF Downloads
8
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...