Explainable AI in the educational domain

JTELSS logo
Explainable AI in the educational domain 📅 Thursday 21/05 10:30-12:00h 📍 Workshop Space A 🔎 Needs Analysis Participants in the educational domain come from diverse backgrounds, including both technical and non-technical profiles. While explainable AI (XAI) methods are increasingly used in learning analytics, there is often a gap between technical

Speakers

Gabriella Casalino
Gabriella Casalino
University of Bari Aldo Moro, Italy

Start

End

Explainable AI in the educational domain

📅 Thursday 21/05 10:30-12:00h
📍 Workshop Space A
🔎 Needs Analysis

Participants in the educational domain come from diverse backgrounds, including both technical and non-technical profiles. While explainable AI (XAI) methods are increasingly used in learning analytics, there is often a gap between technical implementation and practical understanding. Educators and stakeholders need to interpret and use explanations for decision-making, while technical participants need to understand how these explanations are generated. This session addresses the need for a shared understanding of XAI that bridges technical methods and human-centered interpretation.

📒 Session Description

This hands-on session introduces participants to explainable AI techniques in the context of learning analytics. Using a real-world educational dataset, participants will explore how predictive models can be explained using different XAI approaches. The session combines guided demonstrations with interactive group activities, allowing participants to interpret explanations, reflect on their usefulness, and discuss how they can support decision-making in educational settings. Coding is optional: technical participants can explore a Python notebook, while others focus on interpretation and application.

💡 Learning Objectives

By the end of the session, participants will be able to:

  • Understand the role of explainability in learning analytics
  • Interpret common XAI outputs (e.g., feature importance, local explanations)
  • Critically assess the usefulness and limitations of explanations
  • Reflect on how explanations can support decision-making in educational contexts
  • Recognize the importance of human-centered design in XAI
🧩 Pre-activities and prerequisites

No prior programming experience is required. Basic familiarity with AI or data analysis concepts is helpful but not mandatory. Participants with technical backgrounds may benefit from basic knowledge of Python and machine learning. No pre-activities are required.



+ GOOGLE CALENDAR  + ICAL IMPORT