An Integrated Model to Identify Learning Style Using Machine Learning Techniques

Abstract

Identifying the Learning style is important for assisting students in retaining lifelong learned information and improving their understanding of the subject matter in detail. One of the main methodologies for identifying the learning style of the learners is by conducting an analysis based on a questionnaire designed according to a particular learning model. But using a questionnaire may not always be a proper mechanism as this gives an additional burden to the learners and may generate inaccurate results if improper responses are received from the learners. Hence, the most current tendency in classifying the learning style is to identify and use behavioural attributes to automatically detect the learner's learning style without bothering the learner while they are engaged in the learning process. The main intention of the research is to propose an integrated model of Moodle Logs and time spent on academic activities to identify the learning styles of the learners. The proposed model is used to track the student's behaviour during academic activities using Moodle logs and the total time a student spent on each academic activity. To track relevant data, the course content will be designed in Moodle according to the requirement of the Felder-Silverman Learning Style model (FLSM). Data required for the Moodle logs are collected from the logs already available in Moodle, and a plugin will be developed to track the time spent on academic activities. As per the proposed integrated model, a data set will be prepared using both Moodle logs and time tracking. Once the data set is prepared, the machine learning technique is applied to the data to identify the learning style and patterns of the learners. The results show that the integrated model can be used to categorize each student’s learning pattern according to the FLSM. Each student has his own way of using different learning materials. Further, the result shows how the teaching and learning process should be customized for the learners according to the learning pattern of the students.