Forecasting Students' Academic Performance in Educational Data Using Machine Learning Techniques
Saved in:
| Title: | Forecasting Students' Academic Performance in Educational Data Using Machine Learning Techniques |
|---|---|
| Language: | English |
| Authors: | Najah Al-Shanableh (ORCID |
| Source: | International Journal of Information and Communication Technology Education. 2026 22(1). |
| Availability: | IGI Global. 701 East Chocolate Avenue, Hershey, PA 17033. Tel: 866-342-6657; Tel: 717-533-8845; Fax: 717-533-8661; Fax: 717-533-7115; e-mail: journals@igi-global.com; Web site: https://www.igi-global.com/journals/ |
| Peer Reviewed: | Y |
| Page Count: | 23 |
| Publication Date: | 2026 |
| Document Type: | Journal Articles Reports - Research |
| Education Level: | Secondary Education |
| Descriptors: | Artificial Intelligence, Academic Achievement, Prediction, Mathematics Achievement, Secondary School Students, Foreign Countries, Native Language Instruction, Portuguese, Models, Data Analysis, Predictor Variables, Student Characteristics |
| Geographic Terms: | Portugal |
| DOI: | 10.4018/IJICTE.399756 |
| ISSN: | 1550-1876 1550-1337 |
| Abstract: | This study benchmarks multiple machine learning models to predict student academic performance. The research analyzes data from students in mathematics and Portuguese language courses, examining the relationship between various factors and academic performance. The benchmark implementation includes data preprocessing, exploratory data analysis, feature engineering, model training, and hyperparameter tuning for both regression (predicting final grades) and classification (predicting pass/fail outcomes) tasks. The findings demonstrate that ensemble methods, particularly gradient boosting models, outperform other algorithms with root mean square error of 3.34 for regression and F1 score of 0.88 for classification after hyperparameter tuning. Feature importance analysis reveals that past failures, alcohol consumption, study time, and parent education level are among the most influential predictors of academic performance. The results provide valuable insights for educational stakeholders to implement targeted interventions for at-risk students and improve overall academic outcomes. |
| Abstractor: | As Provided |
| Entry Date: | 2026 |
| Accession Number: | EJ1495137 |
| Database: | ERIC |
| Abstract: | This study benchmarks multiple machine learning models to predict student academic performance. The research analyzes data from students in mathematics and Portuguese language courses, examining the relationship between various factors and academic performance. The benchmark implementation includes data preprocessing, exploratory data analysis, feature engineering, model training, and hyperparameter tuning for both regression (predicting final grades) and classification (predicting pass/fail outcomes) tasks. The findings demonstrate that ensemble methods, particularly gradient boosting models, outperform other algorithms with root mean square error of 3.34 for regression and F1 score of 0.88 for classification after hyperparameter tuning. Feature importance analysis reveals that past failures, alcohol consumption, study time, and parent education level are among the most influential predictors of academic performance. The results provide valuable insights for educational stakeholders to implement targeted interventions for at-risk students and improve overall academic outcomes. |
|---|---|
| ISSN: | 1550-1876 1550-1337 |
| DOI: | 10.4018/IJICTE.399756 |
