Multiple Intelligence Learning Style Detection in E-learning using Support Vector Machine (SVM) Algorithm

One of the functions of personalized learning is to provide learning resources/materials that match the needs and preferences of learner characteristics. The management of learner characteristics uses a learner model that contains basic information, knowledge level, learning style, and cognitive traits of the learner. The theory of multiple intelligences is based on cognitive research; it has been applied in traditional learning and successfully influences and impacts learning performance. However, applying multiple intelligence learning styles in e-learning still needs improvement. One of the main obstacles is the low accuracy of this learning style detection in online learning. This research aims to implement a machine learning algorithm, namely Support Vector Machine, which is expected to improve the performance of multiple learning styles classification. This research uses data analysis methods, including data collection, preprocessing, feature selection, model selection, and evaluation. The data in this research is sourced from learners' interactions with an online learning platform, namely a module-based Learning Management System (LMS) owned by the Directorate General of Teachers and Education Personnel, Ministry of Education, Culture, Research and Technology, Republic of Indonesia. This research has produced the results of the Support Vector Machine (SVM)classification accuracy performance test, reaching 83.64 using the Radial kernel, which is promising for future development. © 2024 IEEE.