Utilizing Explainable AI to Enhance Real-time Student Performance Prediction in Educational Serious Games

Serious Games (SGs) enable the collection of valuable student interaction data, allowing for the analysis of student performance through Artificial Intelligence (AI). However, the lack of explainability in AI models represents a significant challenge. This research aims to create an interpretable model capable of predicting students' real-time performance while playing a SG. We developed a model for task completion prediction, resulting in 77.21% balanced accuracy. Additionally, the incorporation of eXplainable AI (XAI) ensures interpretability, supporting personalized learning experiences and unlocking AI benefits for non-technical users.