Open Player Modeling - Using AI to help Reflection and Learning in Serious Games

Open Learner Models (OLMs) traditionally enhance reflection and learning in classrooms. However, as serious games become more popular for teaching complex concepts, traditional OLMs struggle to link in-game actions to educational concepts, limiting opportunities for reflection and learning. Addressing the above, this paper centered around a serious game titled Parallel delves into our approach to designing AI systems to co-relate in-game player actions with learning paradigms associated with Parallel programming. We detail our initial strategies in (a) translating game board states into explainable embeddings that are comprehensible to both humans and machines and (b) using these embeddings, we use loss functions to identify peer board states that best support process reflection and learning. These insights then feed into a visualization system aimed at supporting player reflection and learning.