There is no wealth like Knowledge
                            No Poverty like Ignorance
ARPN Journals

ARPN Journal of Engineering and Applied Sciences >> Call for Papers

ARPN Journal of Engineering and Applied Sciences

Comparative analysis of machine learning techniques and linear regression in predicting engineering student’s performance in differential equation

Full Text Pdf Pdf
Author Melchor G. Pacer
e-ISSN 1819-6608
On Pages 539-546
Volume No. 21
Issue No. 8
Issue Date June 20, 2026
DOI https://doi.org/10.59018/042661
Keywords machine learning, differential equations, linear regression, random forest, support vector regression, educational data mining, engineering education.


Abstract

This study evaluated and compared four predictive models - Linear Regression, Decision Tree, Random Forest, and Support Vector Regression (SVR) - in estimating engineering students' final grades in Differential Equations. Academic grade records of 646 students from the Technological University of the Philippines enrolled in Mathematics in the Modern World (MMW), Calculus 1, Calculus 2, and Differential Equations were used. All models were evaluated using 10-fold cross-validation with R², Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) as performance metrics. Results showed that all models produced near-zero or negative R² values, with Linear Regression performing best (R² = -0.011, RMSE = 2.255, MAE = 1.699). Machine learning models did not outperform the linear baseline, suggesting that prerequisite mathematics grades alone are insufficient predictors of Differential Equation performance. The findings recommend the inclusion of richer predictor variables in future predictive models.

Back

GoogleCustom Search



Seperator
    arpnjournals.com Publishing Policy Review Process Code of Ethics

Copyrights
© 2026 ARPN Publishers