Prediction of Students’ Academic Performance Based on Courses’ Grades Using Deep Neural Networks

Aya Nabil,Mohammed Seyam,Ahmed Abou-Elfetouh

doi:10.1109/access.2021.3119596

Abstract

Predicting students’ academic performance at an early stage of a semester is one of the most crucial research topics in the field of Educational Data Mining (EDM). Students are facing various difficulties in courses like “Programming” and “Data Structures” through undergraduate programs, which is why failure and dropout rates in these courses are high. Therefore, EDM is used to analyze students’ data gathered from various educational settings to predict students’ academic performance, which would help them to achieve better results in their future courses. The main goal of this paper is to explore the efficiency of deep learning in the field of EDM, especially in predicting students’ academic performance, to identify students at risk of failure. A dataset collected from a public 4-year university was used in this study to develop predictive models to predict students’ academic performance of upcoming courses given their grades in the previous courses of the first academic year using a deep neural network (DNN), decision tree, random forest, gradient boosting, logistic regression, support vector classifier, and K-nearest neighbor. In addition, we made a comparison between various resampling methods to solve the imbalanced dataset problem, such as SMOTE, ADASYN, ROS, and SMOTE-ENN. From the experimental results, it is observed that the proposed DNN model can predict students’ performance in a data structure course and can also identify students at risk of failure at an early stage of a semester with an accuracy of 89%, which is higher than models like decision tree, logistic regression, support vector classifier, and K-nearest neighbor.

Highlights

Education plays an important role in the progress of a nation
WORK Educational data mining is an important analytical tool for solving the problem of analyzing the huge amounts of educational data stored in educational settings for the decision-making process, predicting students’ academic performance at an early stage of a semester, and discovering a hidden pattern and significant knowledge from educational data
There are some problems such as the imbalanced dataset in predicting students’ academic performance, which is a serious challenge that leads to poor performance

Summary

Introduction

Education plays an important role in the progress of a nation. It is a crucial tool for success in life. The academic performance of students is an essential factor that influences the accomplishment of any educational institution. During the learning process at different levels of education, the failure rates and dropouts of computer programming courses are two essential problems faced by students [2, 3]. Artificial intelligence (AI) and machine learning (ML) have been applied in various fields such as image classification, natural language processing, speech recognition, text translation, and the field of educational data mining (EDM). EDM is concerned with applying various data mining techniques such as classification, regression, time series analysis, and association rule mining in the education field to analyze and evaluate various aspects of educational datasets collected from different e-learning environments or higher educational institutions. EDM is one of the most common techniques used to develop predictive models to extract hidden patterns and useful information, which can help in education and learning [4]

Objectives

Methods

Results

Conclusion