Skip to main content

Unboxing The Blackbox: New Research on Explainability in AI


Explainable AI for Education (XAI-Ed) is an emerging field of research that aims to develop AI systems that can explain their decision-making process to users in a transparent and interpretable manner. This is particularly important in the context of education, where students and educators need to understand how AI algorithms reach their conclusions in order to improve their learning outcomes.

Unboxing the Blackbox: Learnings from the ALS pilot

Adaptemy are proud to share our most recent research: A joint presentation resulting from our partnership with the Ministry of Education, Singapore. The paper was presented at the 15th International Conference on Education Technology and Computers (ICETC 2023) at University of Barcelona, Spain, by the two lead authors:

  • Head of Research and Learning, Adaptemy: Dr. Ioana Ghergulescu Linkedin
  • Senior Specialist, Technologies for Learning Branch, MoE Singapore: Mr Soo Jiunn Huat Linkedin

Research Abstract

The application of XAI-Ed to an adaptive learning system has the potential to revolutionize the way that students learn. Adaptive learning systems are already being used in classrooms to personalize the learning experience for each individual student, but the lack of transparency in the decision-making process of these systems can be a significant barrier to their effectiveness.

By incorporating Explainable AI into adaptive learning systems, educators and students can gain a deeper understanding of the factors that are influencing the system’s recommendations and can adjust their learning strategies accordingly. For example, if an adaptive learning system suggests that a student should focus more on reading comprehension, the system could explain which specific areas of reading comprehension the student is struggling with and provide tailored resources to help them improve.

Furthermore, Explainable AI can help build trust between students and educators and the AI systems they are using. By providing clear and transparent explanations of how recommendations are being made, students and educators can feel confident that the AI system is a reliable and trustworthy learning companion that is working in their best interest.

This presentation will explicate the application of the XAI-ED framework on an Adaptive Learning System designed by MOE Singapore, and uncover the “black box” in the system with an aim of providing suggestions for further iterations. Overall, the incorporation of XAI-ED into adaptive learning systems has enormous potential to improve the efficacy of education, by enabling personalized and transparent learning experiences for each individual student.

Get In Touch

If your team is committed to improving learning experiences using AI, you can use this link to contact Adaptemy:

Get in Touch