Predicting students' performance based on learning style by using artificial neural networks

Abstract

In recent years, many education researchers concentrated on learning styles and its implications. They relied that people who have different personality types tend to have different learning styles, which in turn influence to students' performance in each type of subjects. To measure the relationship between learning styles and student performance in a subject or entire course, we conducted an online survey with a participation of students in various courses to analysis and show the effects of different learning styles on students' performance. This test based on the Felder-Soloman questionnaire with 44 test questions divided into four dimensions. We have filtered data to select 316 undergraduate students for the analysis. According questionnaire results, we have applied statistical methods to analyze data and issued some important notes. Besides these, we have built an artificial neutral network to predict academic performance based on students' learning style. We have shown some significant relationships between learning style and performance of the testers, which have implications for the studying of a course.