Virtual Booth

Educational institutions often struggle to identify students who need support before their academic performance begins to decline. Although large amounts of student data are collected through learning management systems and academic records, much of this information is underutilized for timely intervention. Existing learning analytics tools mainly provide reports and statistics, offering limited support for predicting risks, understanding underlying causes, or evaluating potential intervention strategies.

To address this challenge, this project introduces  EduTwin, an Educational Digital Twin platform designed to help educators make more informed and proactive decisions. EduTwin creates a digital representation of each student by combining data such as academic performance, attendance, and learning engagement. The platform enables educators to identify students who may be at risk, understand the key factors influencing their performance through explainable AI, and explore different intervention options using an interactive simulation feature.

A key strength of EduTwin is its ability to move beyond prediction. Educators can test possible support strategies, such as mentoring programs or additional tutorials, and view the projected outcomes before implementing them. By combining Educational Digital Twin technology, explainable AI, intervention simulation, and ethical governance in a single platform, EduTwin offers a practical approach to supporting student success. The innovation has the potential to enhance early intervention, improve learning outcomes, and contribute to the development of more inclusive and data-informed educational environments.