<![CDATA[Adapting to students’ learning styles is a key factor in enhancing the effectiveness of higher education, particularly in Informatics programs where learning preferences vary widely. This study aims to segment students based on their learning styles using the K-Means clustering algorithm, guided by the VARK model (Visual, Auditory, Read/Write, Kinesthetic). Data were collected from 130 Informatics students, including information on their learning preferences, and processed through normalization techniques. The optimal number of clusters was determined using the Elbow Method and Silhouette Score, and subsequent cluster interpretation was conducted. The results identified three dominant clusters, each representing distinct learning behavior patterns. These clusters were analyzed to recommend tailored instructional strategies for each group. Specifically, Visual learners were found to benefit from graphic-heavy materials, Auditory learners preferred lectures and discussions, Read/Write learners thrived on written content and detailed notes, while Kinesthetic learners responded best to hands-on activities. The findings support the development of adaptive, data-driven teaching approaches that align with the actual learning tendencies of students in Informatics. Moreover, the study demonstrates that the K-Means method is effective in systematically identifying student learning profiles, which can be used to inform instructional improvements. This personalized approach to teaching could significantly enhance learning outcomes by providing students with the most effective educational experiences tailored to their individual learning styles]]>
