<![CDATA[An online learning system is a very crucial thing nowadays to prevent the spread of COVID-19 virus. However, this system is very difficult to maintain student motivation and engagement because there is no direct interaction between teacher and student. This study reviewed the use of student log data for the needs of learning analytics to predict student performance or drop-out trends from a course by looking at the student interaction log data with the system and student demographic data using open data, namely the Open University Learning Analytics Dataset (OULAD). From reviews of several research articles that refer to these data, we can see: 1) the common problems, i.e., prediction of drop-out student, prediction of student performance and engagement; 2) the features used during modeling, i.e., demographics and interactions, either summarized daily or weekly with various feature representations; 3) learning analysis methods that use machine learning algorithm, i.e., Decision Tree (DT), Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), and Long Short-Term Memory (LSTM). This paper also discusses the risk mitigation process of students, planning and designing data systems that support learning analytics, and problems that are often encountered during the modeling process.]]>
Copyrights © 2020
