Analysis of Mooc Data With Educational Data Mining: Systematic Literature Review

Abstract

Participants’ performance is one of the critical factors for the success of the platforms. There is a lot of data in Massive Open Online Course platforms that are free and open to everyone, and due to this large amount of educational data, it is difficult to make accurate predictions and inferences. The primary purpose of this research is to conduct a literature review to discover the existing Educational Data Mining methods and techniques used to analyze Massive Open Online Course data. For this purpose, the focus is on the source from which the data is collected, which Educational Data Mining methods and techniques are used, and which tools are used in the analysis to compare different approaches. A total of 32 articles published between 2013-2024 were included in the scope of the study. According to the findings, there are many algorithms used for Educational Data Mining methods and techniques in the analysis of Massive Open Online Course data. The most preferred algorithm in the studies is “K-Means”, followed by “Support Vector Machines”, “Decision Trees” and “Random Forest”. Coursera and Edx are among the platforms used and preferred worldwide. It is anticipated that making the data available on these platforms public will contribute to further research and guide studies in the education field. Privacy and ethics also come to the fore within the scope of open data publication. In this context, developing some standards and new approaches to share data with researchers in a standard form that does not include privacy violations will significantly contribute to studies conducted in this field. © 2025, TUBITAK. All rights reserved.