Data Augmentation Approach Using SMOTE Pipeline and Regularized Random Forest for Student Graduation Prediction on a Small-Scale Dataset

Mochamad Fadillah Abdullah

Abstract


Programming courses in computer science curricula often present significant abstraction challenges for first-year undergraduate students, frequently leading to high failure rates and early dropouts. The use of conventional oversampling risks triggering data leakage and overfitting. This study proposes an EDM framework that integrates the Imbalanced-Learn Pipeline (SMOTE), a regularized Random Forest, and Leave-One-Out Cross-Validation (LOOCV) on 27 real student track records in the Algorithm and Programming course. All experimental scenarios were computationally executed using the Google Colaboratory cloud platform. To prevent data leakage, SMOTE was embedded exclusively within the training fold of each LOOCV iteration. Tree depth restriction (max_depth=2) was applied to prevent excessive pattern memorization. The test results demonstrated realistic performance with an overall accuracy of 85.19%, precision of 66.67% (for the target minority/failing class), and recall of 85.71% [1]. The model successfully identified 4 misclassification cases (1 False Positive and 3 False Negatives) attributable to academic anomalies rather than overfitting. Quiz Average (X2) and Structured Practicum Score (X3) were identified as the primary graduation predictors, contributing 29.3% and 28.9% respectively. This methodological combination effectively and scientifically overcomes the Cold-Start Problem and is free from overfitting for early warning system architectures.

Keywords


Educational Data Mining; Cold-Start Problem; SMOTE Pipeline; Random Forest; Leave-One-Out Cross-Validation

Full Text:

PDF

References


E. Ahmed, “Student Performance Prediction Using Machine Learning Algorithms,” Applied Computational Intelligence and Soft Computing, vol. 2024, 2024, doi: 10.1155/2024/4067721.

C. Watson and F. W. B. Li, “Failure rates in introductory programming revisited,” in ITICSE 2014 - Proceedings of the 2014 Innovation and Technology in Computer Science Education Conference, Association for Computing Machinery, 2014, pp. 39–44. doi: 10.1145/2591708.2591749.

Douglas Rakasiwi Nugroho, Nurfiah Idris, Kevin Kurniawan, Devi Fitrianah, Edy Irwansyah, and Gede Putra Kusuma, “Logistic Regression and Random Forest Comparison in Predicting Students’ Qualification Based on Students’ Half-Semester Performance,” IEEE, 2023.

F. F. Anwar, A. I. Jaya, and M. Abu, “Prediksi Kelulusan Mahasiswa Tepat Waktu Menggunakan Metode Decision Tree dengan Penerapan Algoritma C4.5,” JURNAL ILMIAH MATEMATIKA DAN TERAPAN, vol. 19, no. 1, pp. 19–28, Jun. 2022, doi: 10.22487/2540766x.2022.v19.i1.15880.

T. H. Hasibuan and D. Mahdiana, “Prediksi Kelulusan Mahasiswa Tepat Waktu Menggunakan Algoritma C4.5 Pada Uin Syarif Hidayatullah Jakarta,” SKANIKA: Sistem Komputer dan Teknik Informatika, vol. 6, pp. 61–74.

Z. Amri, K. Kusrini, and K. Kusnawi, “Prediksi Tingkat Kelulusan Mahasiswa menggunakan Algoritma Naïve Bayes, Decision Tree, ANN, KNN, dan SVM,” Edumatic: Jurnal Pendidikan Informatika, vol. 7, no. 2, pp. 187–196, Dec. 2023, doi: 10.29408/edumatic.v7i2.18620.

A. K. Darmawan, I. Yudhisari, A. Anwari, and M. Makruf, “Pola Prediksi Kelulusan Siswa Madrasah Aliyah Swasta dengan Support Vector Machine dan Random Forest,” Jurnal Minfo Polgan, vol. 12, no. 2, 2023, doi: 10.33395/jmp.v12i2.12388.

Satrio Junaidi, R. Valicia Anggela, and D. Kariman, “Klasifikasi Metode Data Mining untuk Prediksi Kelulusan Tepat Waktu Mahasiswa dengan Algoritma Naïve Bayes, Random Forest, Support Vector Machine (SVM) dan Artificial Neural Nerwork (ANN),” Journal of Applied Computer Science and Technology, vol. 5, no. 1, pp. 109–119, Jun. 2024, doi: 10.52158/jacost.v5i1.489.

M. Kumar, N. Singh, J. Wadhwa, P. Singh, G. Kumar, and A. Qtaishat, “Utilizing Random Forest and XGBoost Data Mining Algorithms for Anticipating Students’ Academic Performance,” International Journal of Modern Education and Computer Science, vol. 16, no. 2, pp. 29–44, Apr. 2024, doi: 10.5815/ijmecs.2024.02.03.

F. Sihombing, M. P. Parhusip, and K. Frans Samuel Gultom, “CLASSIFICATION OF STUDENTS ACADEMIC ACHIEVEMENT USING A RANDOM FOREST ALGORITHM BASED ON EDUCATIONAL DATA MINING,” Excellent: International Journal of Computational Intelligence and Sustainable Innovation, vol. 1, no. 1, 2026, [Online]. Available: https://jurnal.dharmawangsa.ac.id/index.php/excellent/index

A. S. Yazid, “Eksplorasi Data Akademik untuk Memprediksi Ketepatan Waktu Lulus Mahasiswa Menggunakan Algoritma Naive Bayes,” 2024. [Online]. Available: http://jurnal.mdp.ac.id

F. M. Alnassar, T. Blackwell, E. Homayounvala, and M. Yee-king, “Addressing Class Imbalance in Predicting Student Performance Using SMOTE and GAN Techniques,” Applied Sciences (Switzerland), vol. 16, no. 7, Apr. 2026, doi: 10.3390/app16073274.

A. Anggrawan, H. Hairani, and C. Satria, “Improving SVM Classification Performance on Unbalanced Student Graduation Time Data Using SMOTE,” International Journal of Information and Education Technology, vol. 13, no. 2, pp. 289–295, Feb. 2023, doi: 10.18178/ijiet.2023.13.2.1806.

S. Diantika, H. Nalatissifa, N. Maulidah, R. Supriyadi, and A. Fauzi, “Penerapan Teknik Random Oversampling Untuk Memprediksi Ketepatan Waktu Lulus Menggunakan Algoritma Random Forest,” 2024. [Online]. Available: https://www.kaggle.com/

A. F. Anjani, D. Anggraeni, and I. M. Tirta, “Implementasi Random Forest Menggunakan SMOTE untuk Analisis Sentimen Ulasan Aplikasi Sister for Students UNEJ,” Jurnal Nasional Teknologi dan Sistem Informasi, vol. 9, no. 2, pp. 163–172, Sep. 2023, doi: 10.25077/teknosi.v9i2.2023.163-172.

A. Fauziah and J. Hernadi, “Klasifikasi Data Tak Seimbang Menggunakan Algoritma Random Forest dengan SMOTE dan SMOTE-ENN (Studi Kasus pada Data Stunting) Classification of Imbalanced Data Using Random Forest Algorithm with SMOTE and SMOTE-ENN (Case Study on Stunting Data),” Jurnal Riset Sistem dan Teknologi Informasi (RESTIA, vol. 3, no. 2, pp. 112–121, 2025, [Online]. Available: https://bit.ly/DataSintetis.

L. Breiman, Random Forests, vol. 45. Berkeley: Kluwer Academic Publishers., 2001.

Z. Nurfadilla, “IMPLEMENTASI DATA MINING UNTUK MEMPREDIKSI KELULUSAN MAHASISWA TEPAT WAKTU MENGGUNAKAN RANDOM FOREST,” vol. 1, no. 1, p. 2022, [Online]. Available: https://www.kaggle.com/baladikaalhariri

M. Sulehu, W. Wisda, F. Wanita, and M. Markani, “Optimasi Prediksi Kelulusan Mahasiswa Menggunakan Random Forest untuk Meningkatkan Tingkat Retensi,” Jurnal Minfo Polgan, vol. 13, no. 2, pp. 2364–2374, Jan. 2025, doi: 10.33395/jmp.v13i2.14472.

A. Palanivinayagam and R. Damaševičius, “Effective Handling of Missing Values in Datasets for Classification Using Machine Learning Methods,” Information (Switzerland), vol. 14, no. 2, Feb. 2023, doi: 10.3390/info14020092.

S. Kapoor and A. Narayanan, “Leakage and the reproducibility crisis in machine-learning-based science,” Patterns, vol. 4, no. 9, Sep. 2023, doi: 10.1016/j.patter.2023.100804.

A. Demircioğlu, “Applying oversampling before cross-validation will lead to high bias in radiomics,” Sci. Rep., vol. 14, no. 1, Dec. 2024, doi: 10.1038/s41598-024-62585-z.

J. Gao, D. Lin, and J. Hu, “A Generalizable Seven-Step Workflow for Stacking Ensemble Learning: Data Leakage Auditing, Cross-Institutional Validation, and Negative Findings in Educational Data Mining,” Aug. 06, 2026. doi: 10.21203/rs.3.rs-10290103/v1.

Gareth James, Daniela Witten, Trevor Hastie, and Robert Tibshirani, An Introduction to Statistical Learning with Applications in R. 2021.

S. B. Kotsiantis, “Decision trees: a recent overview,” Artif. Intell. Rev., vol. 39, no. 4, pp. 261–283, 2013, doi: 10.1007/s10462-011-9272-4.

A. Vabalas, E. Gowen, E. Poliakoff, and A. J. Casson, “Machine learning algorithm validation with a limited sample size,” PLoS One, vol. 14, no. 11, Nov. 2019, doi: 10.1371/journal.pone.0224365.

T. L. A. Dinh and F. Aires, “Nested leave-two-out cross-validation for the optimal crop yield model selection,” Geosci. Model Dev., vol. 15, no. 9, pp. 3519–3535, May 2022, doi: 10.5194/gmd-15-3519-2022.

G. W. Cha, H. J. Moon, and Y. C. Kim, “Comparison of random forest and gradient boosting machine models for predicting demolition waste based on small datasets and categorical variables,” Int. J. Environ. Res. Public Health, vol. 18, no. 16, Aug. 2021, doi: 10.3390/ijerph18168530.

G. W. Cha, H. J. Moon, and Y. C. Kim, “Comparison of random forest and gradient boosting machine models for predicting demolition waste based on small datasets and categorical variables,” Int. J. Environ. Res. Public Health, vol. 18, no. 16, Aug. 2021, doi: 10.3390/ijerph18168530.

A. K. Veerasamy, M. J. Laakso, and D. D’Souza, “Formative Assessment Tasks as Indicators of Student Engagement for Predicting At-risk Students in Programming Courses,” Informatics in Education, vol. 21, no. 2, pp. 375–393, 2022, doi: 10.15388/infedu.2022.15.




DOI: http://dx.doi.org/10.38101/sisfotek.v16i2.17254

Refbacks

  • There are currently no refbacks.


 

JURNAL SISFOTEK GLOBAL

Organized by: Research Center and Community Development
Published by: Institut Teknologi dan Bisnis Bina Sarana Global
Jl. Aria Santika No.43A, Margasari, Kec. Karawaci, Kota Tangerang, Banten 15114
Phone. +62 552 2727
Email: lppm@global.ac.id

INDEXED BY:

   


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License..
Based on a work at https://journal.global.ac.id/index.php/sisfotek/index.