Research & Publications
Our commitment to evidence-based learning is supported by extensive academic research and peer-reviewed publications
Online Learning Before, During and After COVID-19: Observations over 20 Years
This paper reflects on two decades of online learning development and implementation, including the rapid changes triggered by the COVID-19 pandemic.
Before COVID-19:
Examines the gradual adoption of online learning, evolving technologies, and challenges faced by educators and institutions.
During COVID-19:
Captures the urgent shift to fully online delivery, infrastructure strain, and pedagogical adaptations for learning continuity.
After COVID-19:
Explores hybrid models, interactive video's growing role, and potential for scalable, flexible learning experiences.
Video-Based Digital Learning Framework Using AI Question Generation and Answer Assessment
Co-authors: Abdue Rahim Mohammad Forkan, Yong-Bin Kang, Prem Prakash Jayaraman, Hung Du, Steven Thomson, Liz Kollias, Natalie Wieland
This paper proposes a framework that combines video-based learning with AI-driven automated question generation and answer assessment.
- Details how AI algorithms can create questions directly from video transcripts and assess learner responses
- Aims to increase engagement, improve knowledge retention, and allow large-scale personalised assessment
- Includes discussion of technical architecture, data flow, and LMS integration
- Addresses both scalability for mass learning environments and customisation for targeted skill areas
Comparative Analysis of Large Language Models for Automated Question Generation from Video-Based Learning Content
Co-authors: Yong-Bin Kang, Abdur Forkan, Abhik Banerjee, Prem Prakash Jayaraman, Anthony McCosker, Sungsoo Kim, Natalie Wieland, Liz Kollias
This study evaluates different large language models (LLMs) for generating educational questions from video content.
- Compares models on accuracy, relevance, cognitive depth of generated questions, and adaptability to various subject domains
- Includes performance benchmarks, error patterns, and recommendations for optimal model selection
- Discusses practical applications including automated quiz generation for MOOCs, professional training modules, and interactive compliance learning
- Highlights trade-offs between cutting-edge generative performance and computational efficiency for real-world deployment