22nd AIAI 2026, 16 - 19 July 2026, Chania, Crete, Greece

GenAI-Powered Feedback for Video-Based Learning

Ordóñez Josselyn, Anaya Antonio, Zarauz Monero Antonio, G. Zacchigna Federico

Abstract:

  As virtual higher education expands, providing scalable and actionable pedagogical feedback remains a significant challenge. This paper presents an automated system developed for the Laboratory of Embedded Systems to evaluate and enhance teaching quality in postgraduate courses. Our methodology integrates multimodal audio analysis with an iterative prompt engineering process, validated against student satisfaction surveys. The research identified key pedagogical attributes that demonstrate significant Pearson correlations with student ratings. The system generates structured, evidence-based feedback that includes numerical scores and qualitative justifications in natural language. This work establishes a robust, data-driven framework for continuous instructional improvement, bridging the gap between raw educational recordings and actionable pedagogical insights in digital learning environments.  

*** Title, author list and abstract as submitted during Camera-Ready version delivery. Small changes that may have occurred during processing by Springer may not appear in this window.