Research Portfolio

AIEDU Lab conducts research on AI-supported learning environments, self-regulated learning, educational technology, learning analytics, writing support, assessment, and responsible uses of generative AI in education. Our work combines learning sciences, empirical educational research, and system development to design and evaluate AI-supported tools for inclusive, learner-centred education.

The lab organizes its current and emerging work around the following research clusters.

MetaMentorAI and AI-Supported Self-Regulated Learning

Status: Active development

PI: Dr. Michael Lin, Dr. Jeeho Ryoo; Collaborators: Daniel Chang; Student Assistants: Alex Yang, Alex Jing-Yuan Huang

MetaMentorAI is a multi-institutional project focused on designing an AI-supported learning environment that helps students monitor, reflect on, and regulate their learning. The platform integrates learning analytics, metacognitive prompts, knowledge visualization, and conversational feedback to support deeper engagement and self-regulated learning.

This project builds on research in self-regulated learning and technology-enhanced learning environments. It explores how AI can provide timely guidance while preserving learner agency, reflection, and meaningful engagement with learning materials.

Current directions: AI-supported metacognitive guidance; learner interaction analytics; semantic knowledge graphs; conversational feedback loops; evaluation of learner engagement and regulation.

AI-Supported Writing, Argumentation, and Revision

Status: Active research and tool development

This research cluster connects two related strands of work: evidence synthesis on argument diagram research and the development of AI-supported tools for writing revision.

Review of Argument Diagram Research
PI: Daniel Chang; Collaborators: Dr. Michael Lin, Dr. Jeeho Ryoo, Dr. Gwo-Jen Hwang, Dr. Vivien Lin

This project reviews studies on the use of argument diagram tools to enhance writing, reasoning, and learning. The goal is to identify evidence gaps, compare tool features, and examine instructional strategies that can guide future research and pedagogy.

Milestones: Scoping review completed; meta-analysis submitted to a journal; systematic review in progress.

Re-Mapping Tool: Using Concept Maps to Support Revision Activity for Writing
PI: Daniel Chang, Dr. Jeeho Ryoo, Dr. Michael Lin

Building on prior review work on argument diagrams and writing support, this project develops Re-Map, an AI-supported revision tool that helps students reflect on the structure and clarity of their writing. Students upload draft essays to the platform, and the system generates concept-map-based representations of their ideas. These representations are paired with guiding questions to help students identify relationships among concepts, reconsider organization, and make more intentional revisions.

Current directions: Concept-map-based writing feedback; AI-supported revision; student reflection during writing; evaluation of revision quality and learner engagement.

Generative AI in Computer Science Education and Educational Practice

Status: Evidence synthesis and publication development

PI: Dr. Michael Lin; Collaborators: Daniel Chang, Dr. Jeeho Ryoo

This project maps empirical research on how generative AI is being used in computer science education and related learning contexts. The work examines how generative AI supports learning, programming practice, skill development, feedback, and instructional design.

The goal is to identify patterns in the current literature, clarify what forms of evidence are available, and inform more responsible and evidence-based integration of generative AI into teaching and learning.

Milestones: Rapid review completed.

Current directions: Systematic review methods; empirical evidence mapping; AI-supported programming education; instructional opportunities and risks in generative AI use.

Emotion-Aware AI and Self-Regulated Learning

Status: Emerging research direction

This emerging line of work examines how AI-supported learning systems might better recognize and respond to the emotional dimensions of learning. Students’ engagement with AI is not only cognitive; it also involves uncertainty, frustration, confidence, motivation, and help-seeking.

The project explores how AI could support emotion regulation within self-regulated learning without over-automating the learning process or replacing students’ own reflective judgement.

Current directions: Emotion regulation in SRL; affect-sensitive AI feedback; learner uncertainty and help-seeking; ethical limits of emotional inference in AI-supported learning systems.

Agentic AI, Gamification, and Writing Development

Status: Concept and prototype development

This project explores how agentic AI and gamified learning design can support students’ English writing development. The proposed platform uses multiple AI agents as writing mentors that provide tiered feedback on clarity, style, organization, and narrative coherence.

The project focuses on iterative revision. Instead of using generative AI simply to produce text, the system is designed to encourage students to revise, compare alternatives, and develop greater control over their own writing choices.

Current directions: Agentic AI feedback; writing revision; gamified motivation; English writing development; learner engagement with AI-generated feedback.

Responsible Generative AI, Assessment, and Evidence of Learning

Status: Emerging research and grant development area

This cluster examines how educators and institutions can design assessment practices that respond to generative AI while preserving meaningful evidence of student learning. Rather than treating AI only as a problem of misconduct, this work investigates how assessment, feedback, disclosure, and learning design can be redesigned for AI-rich educational contexts.

This direction connects to teacher education, academic integrity, AI literacy, and the design of learning activities that make student thinking, process, and reflection more visible.

Current directions: AI disclosure practices; assessment redesign; evidence of learning; teacher AI literacy; responsible classroom integration of generative AI.

For student, collaborator, or institutional partnership inquiries, please visit the Join Us or Contact page.