E-learning and COVID-19: Predicting Student Academic Performance Using Data Mining Algorithms

The satisfaction of E-learners has the main effect on the success of the E-learning process and leads to improvements in the E-learning system's quality and several factors affect this satisfaction. Based on the dimensions of e-learning, the main objective of this study was to evaluate the factors that contributed to students' satisfaction with e-learning during pandemic the Covid-19 and to give a thorough understanding and knowledge of different data mining techniques that have been used to predict student performance and development, as well as how these techniques help in the identification of the most relevant student attribute for prediction. Currently, to search for information in large databases, data mining techniques have become very popular and proven itis effective. Because of the performance and effectiveness of data mining techniques, it has been adopted by many areas such as telecommunication, education, sales management, banking, etc. In this paper, data mining algorithms were relied on to build e-learning classification models for a "student performance" data set, the proposed model includes 1000 instances with 35 attributes. Data mining algorithms have been implemented on the student performance data set in E-learning. Among these algorithms are the Decision Tree algorithm, Random Tree algorithm, Naive Bayes algorithm, Random Forest algorithm, REP Tree algorithm, Bagging algorithm and KNN algorithm. After comparing the results and conducting the assessment, the impact of the proposed features in e-learning on the student's performance was clarified. The final result of this study is important for providing greater insight into evaluating student performance in the COVID-19 pandemic and underscores the importance of data mining in education.