What question did this study set out to answer?

This research aims to improve dropout prediction for students in higher education using a machine learning approach.

June 10, 2026Scientific ReportsOpen Access

SOD-FE: a supervised outlier detection and feature engineering approach for student dropout prediction

View Full Paper

Ask AI

Bookmark

This research aims to improve dropout prediction for students in higher education using a machine learning approach.
Developed SOD-FE, a supervised machine learning model employing outlier detection and feature engineering techniques.
Utilized interquartile range outlier detection, mutual information, and Pearson correlation for data refinement.
Evaluated the model using two datasets with five classification algorithms, primarily focusing on Random Forest and Extreme Gradient Boosting.
Achieved F1 scores of 98.09% and 98.33% for Portugal and Slovakia datasets, respectively.
SOD-FE outperformed baseline Random Forest models by 7.46% and 3.37% in F1 score for the two datasets.
The method demonstrated competitive results that align with or exceed the highest performances reported in literature by up to 7.33%.

No takes yet. Share an insight, caveat, or question.

Randomized trial demonstrates improved dropout prediction in students, suggesting enhanced early intervention strategies.

Alzabidi et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark