Projects
Publically funded projects: JSPS KAKEN, research-er.jp
MetaStep: Large Language Model-based Metacognitive Scaffolding in Introductory Programming EducationThis project aims to develop MetaStep, an LLM-based support system designed to promote self-regulated learning in introductory programming education. By seamlessly integrating students’ cognitive and metacognitive models into LLM-based scaffolding, MetaStep enables a gradual transfer of regulatory control from AI to students, supporting their development into self-directed learners. ![]() |
GazeAgency: Utilising Dynamic AOI tracking and Machine Learning for Investigating Learner Agency in AI-Mediated Online Learning EnvironmentsThis Japan–India collaborative project investigates dynamic AOI tracking and machine learning as methods for understanding learner agency in AI-mediated online learning environments. Integrating gaze analysis, behavioural pattern detection, and NLP-based descriptions, the project will extend existing analytical methods and evaluate their applicability within Kyushu University’s AI-supported learning environment. It aims to advance methodological approaches for investigating learner agency, generate preliminary findings and strengthen international collaboration in AI-supported education and learning analytics. |
miXai^learn: Multimodal Interactions, and Explainable AI for Embodied learningThis joint research takes the standpoint that learning has embodied experiences distributed in the physical and virtual spaces. While current online learning systems often focus on the cognitive skills, there is a lack of data-driven approaches for tracking learning process and provide feedback on tasks that are distributed in physical and virtual spaces. This research explores an innovative integration of Artificial Intelligence (AI) and socially assistive robotics in a technical infrastructure to enhance learner’s critical reflection in the psychomotor and cognitive aspects of learning. ![]() |
NEXT-AIR: International Research Network for Next-Generation Learning Support through AI and RoboticsThis project is an international collaboration between Kyushu University and National Taiwan University to explore how AI and robotics can support next-generation education. It brings together experts, young researchers, and graduate students to develop multimodal learning analytics, design adaptive intervention models, and formulate human-centered design principles for next-generation embodied learning. ![]() |
MMLAIR: Multimodal Learning Analytics in ReadingThis project analyzes students' behavioral, cognitive, and emotional patterns during digital reading to understand learning processes and predict knowledge outcomes. By continuously monitoring multimodal data from learning logs, eye-tracking, and physiological sensors, the research enables adaptive learning systems that respond dynamically to individual student states, supporting personalized educational interventions in real-time. ![]() |
GOAL project: AI-supported self-directed learning lifestyle in data-rich educational ecosystemLearners need to be trained to be self-directed in their learning and daily activities. This research focuses on developing digital learning environments that trace learners' behaviors and provide them functionalities to develop such self-direction skills by analysis of those data. ![]() |
Cyberspace Infrastructure Technology Using Big Data and AI: Research on Evidence-Based Tailor-Made EducationIn this research, we collect and analyze learning data logs which is otherwise difficult to synthesize automatically and requires advanced interaction tracking. The approach consumes the domain knowledge of learning and cognitive science, artificial intelligence, and information and communication infrastructure technology. By researching and developing the AI driven pedagogical platform, we aim to realize a precision education that can be tailored to the characteristics of learners based on evidence. ![]() |





