تحليل الأداء الأكاديمي للطلاب باستخدام تقنيات تعلم الآلة

Abstract EN

Predicting students ’academic performance is vital to the success of any education system.

Attracting students and the way to preserve and elevate them is an essential part of that system.

As it affects the university rank, the educational reputation of the institution, and financial support for it.

Monitoring student performance and how to maintain and improve them has become one of the most important priorities for decision-makers in higher education institutions.

The system of improvement and how to maintain students begins with a comprehensive understanding of the reasons behind their dropping out and delinquency from the educational process.

Such an understanding is the basis for accurate prediction of students at risk and thus appropriate intervention to ensure that they are retained.

Many data mining techniques are used such as aggregation, classification, regression, and prediction to help decision-makers accomplish their tasks.

In this study, a new student performance prediction model was introduced based on features that have a major impact on academic achievement.

This study has relied on its mechanism of work on some machine learning algorithms using data collected from the Kalboard 360 e-learning system.

The expert system achieved after several practical experiments, analysis, and comparison of several machine learning algorithms with an accuracy of 85.2%.