Detecting learning styles in learning management systems using data mining

Abstract

The use of data mining in the education sector has increased in the recent past. One reason for this is the wide use of learning management systems (LMS), which store data related to learning activities. The goal of this research is to predict individual learning styles using the Moodle LMS by analyzing log data using a data mining technique. We use the Waikato environment for knowledge analysis (Weka), as the data mining tool and compare the differences in the performance of several data mining techniques using course log data. Our experimental results show that the J48 decision tree classification algorithm works best with our dataset. We also propose a group learning map that visualizes the learning styles in a class, which can help instructors and learners achieve learning outcomes more effectively.