COMPARISON OF NAIVE BAYES, DECISION TREE, AND RANDOM FOREST ALGORITHMS IN CLASSIFYING LEARNING STYLES OF UNIVERSITAS KRISTEN INDONESIA TORAJA STUDENTS

Abstract

Learning style is an individual's habit or way of absorbing, processing, and managing information. This factor is very important in achieving learning goals. However, in reality, learning styles are often overlooked in the learning process, which can lead to suboptimal absorption of lessons and affect the quality of education. Various models have been developed by educational experts to identify students' learning styles, one of which is the VAK model (Visualization Auditory Kinesthetic) for grouping learning styles. This study compares algorithms in classifying learning styles using the VAK model. The results showed that the most dominant learning style was kinesthetic with a percentage of 46.9% or 478 students. The algorithm modeling showed that Naive Bayes had the highest accuracy with a value of 75%, while Random Forest had the lowest accuracy with a value of 59%. This suggests that Naive Bayes is more suitable for classifying students' learning styles. In conclusion, understanding students' learning styles is crucial for effective education. The VAK model is one way to identify learning styles, and Naive Bayes is a suitable algorithm for classifying students' learning styles. By considering learning styles, educators can tailor their teaching methods to better suit their students' needs and improve the quality of education.