ASMIRES´26 Advanced Statistical Methods for Intelligent and Reliable E-Learning Systems

​

Session Organizer and Chair: Teresa A. Oliveira

Co-Chairs: Amílcar Oliveira and Elisa Henning

 

As digital education increasingly integrates artificial intelligence, learning analytics, and generative AI tools, e-learning platforms are evolving into complex intelligent systems. Ensuring that these systems are reliable, robust, fair, and pedagogically sound, requires strong statistical foundations and rigorous validation methodologies.

Organized by members of LE@D – Laboratory of Distance Education and eLearning at Universidade Aberta (UAb), Portugal, this session explores advanced statistical approaches that strengthen the reliability and quality of AI-enhanced e-learning systems. The focus includes experimental design methodologies, statistical validation frameworks, quality control mechanisms, risk analysis, and techniques for addressing imbalanced educational data in learning analytics.

Particular attention will be given to the application of design of experiments (DoE), balanced incomplete block designs (BIBDs), statistical quality control, and robust model evaluation strategies for adaptive learning environments and generative AI tools used in education. The session also examines reliability and risk mitigation challenges related to automated feedback systems, bias detection, transparency, and responsible AI integration in digital education.  By bridging statistical science, artificial intelligence, and educational innovation, this session contributes to the development of trustworthy, data-driven e-learning ecosystems aligned with ICITED’s commitment to advancing innovative and evidence-based educational practices.

 

Session Objectives

• Present advanced statistical methodologies for validating intelligent e-learning systems.

• Discuss experimental design strategies for optimizing AI-driven educational technologies.

• Explore statistical quality control and risk mitigation frameworks in digital education.

• Address reliability, fairness, and transparency challenges in generative AI for education.

• Strengthen interdisciplinary collaboration within the global e-learning research community.

 

Keywords

E-learning, Learning Analytics, Generative AI, Experimental Design, Statistical Validation, Risk Analysis, Quality Control, Intelligent Educational Systems, Distance Education

​

​

​Contacts for further information: Teresa.Oliveira@uab.pt

​

​