Enhancing Student Adaptability in Online Education through Federated Learning and Explainable AI

In the present digital era, the transition to online education requires the implementation of effective strategies and instruments to anticipate and facilitate student adaptability. This study suggests the creation of an Explainable AI (XAI) model that is based on federated learning. The objective of this model is to predict the adaptability of students to online education across multiple institutions, while also ensuring data privacy, security, and transparency. Federated learning facilitates the decentralized training of machine learning models, thereby enabling institutions to collaborate without the exchange of sensitive data. The interpretability of the model is improved by the incorporation of XAI techniques, such as SHAP (Shapley Additive Explanations) and LIME (Local Interpretable Model-agnostic Explanations), which offer educators transparent insights into the factors that influence student adaptability.