Predictive Modeling of Global Educational Outcomes: A Comparative

Analysis Using Machine Learning Regression Techniques

Abdelhameed Ibrahim 1 ∗, Abdelaziz A. Abdelhamid2, Ehab M. Almetwally3

1Computer Engineering and Control Systems Department, Faculty of Engineering, Mansoura University,

Mansoura 35516, Egypt

2Department of Computer Science, Faculty of Computer and Information Sciences, Ain Shams University,

Cairo 11566, Egypt

3Department of Mathematics and Statistics, Faculty of Science, Imam Mohammad Ibn Saud Islamic

University (IMSIU), Riyadh 11432, Saudi Arabia

Emails: afai79@mans.edu.eg, abdelaziz@cis.asu.edu.eg, emalmetwally@imamu.edu.sa

Abstract

Education contributes a crucial portion to the world’s development; thus, it is crucial to focus on education

enrollment and quality education. It is essential not only that children enroll in school but also that they receive

proper education to improve individuals and, consequently, society. This paper aims to use machine learning

to predict educational outcomes based on the World Educational Data obtained from Kaggle to analyze the

data, preprocess it, and evaluate the performances of the different regression models. The following models

consist of Support Vector Regression (SVR), CatBoost, RandomForestRegressor, ExtraTreesRegressor,

XGBoost, MLPRegressor, GradientBoostingRegressor, DecisionTreeRegressor, KNeighborsRegressor, LinearRegression,

and Pipeline. Evaluation measures used included MSE, RMSE, MAE, MBE, r, R2, NSE, and

WI. Analyzing the performance comparison, the best accuracy was associated with CatBoost with an r value

equal to 0.999996 and an R2 value of 0. 999993; The MSE score was 0.04024. The outcomes of the present

paper demonstrate that the application of advanced machine learning algorithms can be used effectively to

predict educational outcomes, thus enabling policymakers and educational planners to use them for designing

effective educational policies and overcoming existing global challenges in the sphere of education.

Keywords: Educational Data Analysis, Regression Models, Machine Learning, Predictive Modeling, Global

Education Outcomes