Personalized Learning Style System Based on Student Activity Segmentation in Learning Management System Using the Felder Silverman Model with Naïve Bayes

Personalized ways of learning makes it easier for students to learn with their own styles. Through this research, it is expected to produce a learning style recommendation system obtained from the process of student activities carried out in the Learning Management System (LMS). The data used is log data that was obtained from several classes in College X. These classes are selected based on the sufficient variety of learning content carried out in the LMS. The data is processed in the Felder-Silverman model which produces a personalized learning style which is later also processed using naïve bayes methods. The learning style data that was obtained then compared with the results of the learning style index. In this research, the processed data undergoes testing using the Naïve Bayes method, yielding results indicating that Naïve Bayes achieved an average percentage result exceeding 90%.