Predicting Student Performance Using Data Mining Technology in E-Learning: A Review

Education data mining is analyzing educational data to improve decision-making and learning outcomes that are important for educational organizations. Utilizing educational data mining techniques is important for improving modern education. During the COVID-19 pandemic, E-learning has become more prevalent, and predicting student performance in this context has become a significant challenge. Studying and analyzing educational data is important, especially when predicting student performance. There are several factors theoretically assumed to affect student performance. These factors include the student’s prior academic achievement, study habits, access to resources, quality of teaching, class size, and the learning environment. Additionally, factors such as student engagement, support services, and institutional policies can also have an impact on student performance. After surveying, we found that random forest (RF) and recurrent neural networks (RNN) are the best models to classify and predict student performance. They perform well on educational data and can be efficiently used to predict student performance and in early warning systems.