The recommendation of courses in an online educational system is done using different factors that include course length, duration, and learning material information. The course design is based on the instructor's teaching style, which may not necessarily correspond with the various learning preferences of students. Matching the learning style of the student with that of the course is significant for course selection. To address this issue, an intelligent automated mechanism is needed to categorize the course learning style. This research paper presents a course learning style identification model that is based on the Felder-Silverman Learning Style Model (FSLSM). This model will facilitate teachers and learners in designing and choosing a course compatible with their learning style. The model has been designed to align the online course structures with various learning styles, particularly in computer science education. By integrating time-related considerations, this model serves instructional designers, educators, and students, contributing to improved course relevance and learning outcomes. Ultimately, this work fosters the development of more personalized and effective e-learning environments. The model has been evaluated through an experimental study, whose results are employed to further tune the model to improve its accuracy. Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) evaluation metrics have been used to evaluate the results, which demonstrate the effective categorization of course learning style.