Classification of Student Learning Styles Using the Naïve Bayes Algorithm

Abstract

This research aims to classify student learning styles (Visual and Auditory) using the Naïve Bayes algorithm. Data from 100 students were collected with features including 'Minat Visual' (Visual Interest), 'Minat Auditori' (Auditory Interest), and 'the Highest Score' (Highest Score). Categorical data were converted into numerical form through Label Encoding. The dataset was then split into 80% training data and 20% testing data. The Naïve Bayes model, specifically CategoricalNB, was trained using the training data and evaluated using accuracy and Confusion Matrix. The results show that the model achieved an accuracy of 80% on the test data. The Confusion Matrix indicates that out of 10 students who were actually Auditory, 7 were correctly classified, while 3 were misclassified as Visual. Out of 10 students who were actually Visual, 9 were correctly classified, while 1 was misclassified as Auditory. The distribution of learning styles reveals that 52.5% of students tend to have a Visual learning style and 47.5% tend to be Auditory. This research is expected to provide a better understanding of student learning style profiles to enhance learning effectiveness.