The rapid development of Artificial Intelligence and Neural Networks has opened new opportunities to understand and support human learning. Modern educational environments generate large amounts of multimodal and dynamic data from learners’ activities and physiological and behavioral signals, which serve as predictors of learners' cognitive and affective states and can be modeled using neural and hybrid approaches to build intelligent, adaptive, and personalized learning systems. This Special Session aims to bring together researchers working at the intersection of neural computation, learning analytics, cognitive modeling, and educational technologies. Contributions are invited on novel deep learning architectures, neuro-symbolic systems, and hybrid models that can analyze, predict, and enhance learning processes in formal and informal contexts. Relevant topics also include explainable and trustworthy educational AI, affective computing, student modeling, personalization strategies, and ethical considerations in the design of human-centered educational systems.
Possible topics related to application in the healthcare domain, include (but are not limited to):
Neural and hybrid architectures for learning analytics and adaptive education
Deep learning for learner modeling and performance prediction
Cognitive and affective modeling in educational contexts
Personalization and recommendation systems in intelligent tutoring
Neuro-symbolic and hybrid reasoning for educational data
Explainability, fairness, and trust in educational AI
Multimodal learning data and temporal modeling
Human-centered design and evaluation of educational AI systems