<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>Predicting students’ academic performance is a critical research area, yet imbalanced educational datasets, characterized by unequal academic-level representation, present challenges for classifiers. While prior research has addressed the imbalance in binary-class datasets, this study focuses on multi-class datasets. A comparison of ten resampling methods (SMOTE, Adasyn, Distance SMOTE, BorderLineSMOTE, KmeansSMOTE, SVMSMOTE, LN SMOTE, MWSMOTE, Safe Level SMOTE, and SMOTETomek) is conducted alongside nine classification models: K-Nearest Neighbors (KNN), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Support Vector Machine (SVM), Logistic Regression (LR), Extra Tree (ET), Random Forest (RT), Extreme Gradient Boosting (XGB), and Ada Boost (AdaB). Following a rigorous evaluation, including hyperparameter tuning and 10 fold cross-validations, KNN with SmoteTomek attains the highest accuracy of 83.7%, as demonstrated through an ablation study. These results emphasize SMOTETomek’s effectiveness in mitigating class imbalance in educational datasets and highlight KNN’s potential as an educational data mining classifier.</p>
</abstract>

Predicting students’ academic performance is a critical research area, yet imbalanced educational datasets, characterized by unequal academic-level representation, present challenges for classifiers. While prior research has addressed the imbalance in binary-class datasets, this study focuses on multi-class datasets. A comparison of ten resampling methods (SMOTE, Adasyn, Distance SMOTE, BorderLineSMOTE, KmeansSMOTE, SVMSMOTE, LN SMOTE, MWSMOTE, Safe Level SMOTE, and SMOTETomek) is conducted alongside nine classification models: K-Nearest Neighbors (KNN), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Support Vector Machine (SVM), Logistic Regression (LR), Extra Tree (ET), Random Forest (RT), Extreme Gradient Boosting (XGB), and Ada Boost (AdaB). Following a rigorous evaluation, including hyperparameter tuning and 10 fold cross-validations, KNN with SmoteTomek attains the highest accuracy of 83.7%, as demonstrated through an ablation study. These results emphasize SMOTETomek’s effectiveness in mitigating class imbalance in educational datasets and highlight KNN’s potential as an educational data mining classifier.

Comparing Different Oversampling Methods in Predicting Multi-Class Educational Datasets Using Machine Learning Techniques

Cybernetics and Information Technologies

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Predicting students’ academic performance is a critical research area, yet imbalanced educational datasets, characterized by unequal academic-level...

Comparing Different Oversampling Methods in Predicting Multi-Class Educational Datasets Using Machine Learning Techniques

Published Online: Nov 30, 2023

Page range: 199 - 212

Received: Oct 11, 2023

Accepted: Nov 17, 2023

DOI: https://doi.org/10.2478/cait-2023-0044

Keywords
Imbalance educational datasets, Students’ academic performance, Educational data mining, Data re-sampling

© 2023 Muhammad Arham Tariq et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Comparing Different Oversampling Methods in Predicting Multi-Class Educational Datasets Using Machine Learning Techniques

Published Online: Nov 30, 2023

Page range: 199 - 212

Received: Oct 11, 2023

Accepted: Nov 17, 2023

DOI: https://doi.org/10.2478/cait-2023-0044

KeywordsImbalance educational datasets, Students’ academic performance, Educational data mining, Data re-sampling

© 2023 Muhammad Arham Tariq et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Keywords
Imbalance educational datasets, Students’ academic performance, Educational data mining, Data re-sampling