This study explored the processing of active and reflective learning styles among learners and their identification in a mobile learning (m-learning) environment. Leveraging various contextual cues and learning traces, we employed an innovative approach to adapt educational resources based not only on individual learning styles but also on immediate contextual factors, such as schedule, location, and Internet connectivity. Initially, we used unsupervised clustering to classify learners according to their preferences in the processing dimension of Felder and Silverman’s learning-style model. Subsequently, we developed a model for adapting to learning styles, favoring the decision tree algorithm because of its high accuracy in identifying the learning styles. Additionally, we introduced an approach to adapt learning resources based on learners’ styles and contextual characteristics, achieved through the integration of contextual features into the decision tree model’s decision-making process. The empirical results demonstrated the effectiveness of our approach in enhancing the adaptability of educational systems.