<![CDATA[Identifying students’ learning styles is an important factor in supporting adaptive and data-driven learning. However, conventional methods based on manual questionnaires still have limitations in terms of efficiency and accuracy for data processing. This study presents a comparative analysis of machine learning algorithms to classify student learning styles based on questionnaire data. The dataset used consists of 1,170 student data with three learning style classes, namely visual, auditory, and kinesthetic. The four supervised learning algorithms used are Naïve Bayes, Decision Tree, Random Forest, and K-Nearest Neighbors. Model performance evaluation was conducted using 5-fold (80:20) and 10-fold (90:10) cross-validation with accuracy, precision, recall, and F1-score metrics. The results of the experiment show that the Naïve Bayes algorithm has the most optimal and stable performance with the highest accuracy value of 90.60% in both validation scenarios. These findings indicate that machine learning-based classification approaches, particularly Naïve Bayes, are effective for identifying student learning styles and have the potential to support the development of adaptive and personalized learning systems.]]>
Copyrights © 2026
