Estimation of the Academic Performance of Students in Distance Education Using Data Mining Methods

Abstract

Many institutions in the field of education have been involved in distance education with the learning management system. In this context, there has been a rapid increase in data in the e-learning process as a result of the development of technology and the widespread use of the internet. This increase is in the size of large data. Today, big data can be primarily processed, the relationships between data can be discovered, a meaningful conclusion can be drawn, and predictions about the future using big data can be made. However, these data are generally not used in a way to contribute to the people and institutions (educators, education administrators, ministries, etc.) involved in the education process. Therefore, this study aims to estimate the academic success of students who receive education in the distance education process using data mining methods. The reason why data mining is used is that these methods are particularly effective and powerful tools in classification and prediction processes. The methods used in the study are Random Forest, Artificial Neural Networks, Naive Bayes, Support Vector Machines, Logistic Regression, and Deep Learning algorithms, respectively. The dataset includes primary, secondary, and high school students' data, which were obtained from the learning management system used in the distance education process. As a result, the study findings showed that Deep Learning, Random Forest, and Support Vector Machines algorithms provide prediction success at higher performance than others.

Prediction of Student’s Academic Performance through Data Mining Approach

“…The authors [21] used 7 different techniques including Naive Bayes, Random Forest, Support Vector Machine, K-Nearest Neighbour, Logistic Regression, Artificial Neural Network and Deep Learning to examine learning management system data of distance learning programs. The authors used a prediction model based on three categories low, medium and high.…”

Research on Student Behavior Analysis and Grade Prediction System Based on Student Behavior Characteristics

“…Trakunphutthirak et al used a progressive temporal data mining method of educational information on the basis of data mining technology to effectively improve the prediction accuracy of students' academic performance [3]. To effectively evaluate the learning achievements of students in online distance education, Bütüner et al conducted in-depth research on artificial neural networks and deep learning algorithms in data mining methods, so as to effectively improve the prediction accuracy of students' learning achievements [4]. Shreem et al have improved on the basis of genetic algorithm to achieve effective mining of educational data and effective prediction of student performance, thus effectively realizing the prediction classification of student learning performance [5].…”