Prediksi Risiko Performa Akademik Siswa Menggunakan Random Forest, XGBoost, CatBoost, SMOTENC dan Explainable AI

Giri Reksa Guritno; ANGGER SHALALUDDIN AL AYYUBI; Gandung Triyono

doi:10.30865/json.v8i1.10038

Authors

Giri Reksa Guritno Universitas Budi Luhur
Angger Shalaluddin Al Ayyubi Universitas Budi Luhur
Gandung Triyono Universitas Budi Luhur

DOI:

https://doi.org/10.30865/json.v8i1.10038

Keywords:

Performa Akademik, CatBoost, Explainable AI, Imbalanced data, SMOTENC

Abstract

Prediksi risiko performa akademik diperlukan untuk membantu institusi pendidikan mengidentifikasi siswa yang membutuhkan intervensi lebih awal. Tujuan penelitian ini adalah mengevaluasi pengaruh penanganan data tidak seimbang dan interpretabilitas model terhadap prediksi risiko performa akademik siswa. Penelitian ini menggunakan dataset sekunder Student Performance Factors berjumlah 6.377 data dengan target biner Berisiko dan Tidak Berisiko. Penelitian membandingkan Random Forest, XGBoost, dan CatBoost pada dua skenario, yaitu tanpa SMOTENC dan dengan SMOTENC, serta menambahkan explainable AI melalui feature importance, SHAP, dan analisis arah pengaruh fitur. Evaluasi dilakukan menggunakan accuracy, precision macro, recall macro, F1 macro, F1 weighted, recall per kelas, confusion matrix, stratified 5-fold cross-validation, dan uji signifikansi. Hasil pengujian menunjukkan bahwa CatBoost dengan SMOTENC memperoleh performa terbaik dengan accuracy 94,83%, precision macro 93,06%, recall macro 93,06%, dan F1 macro 93,06%. Penerapan SMOTENC meningkatkan recall kelas Tidak Berisiko pada CatBoost dari 85,76% menjadi 89,56%. Hasil interpretasi menunjukkan bahwa Attendance, Hours_Studied, Tutoring_Sessions, Access_to_Resources, dan Previous_Scores merupakan faktor dominan, dengan nilai yang lebih tinggi cenderung menurunkan prediksi Berisiko. Temuan ini menunjukkan bahwa kombinasi ensemble learning, penanganan imbalanced data, validasi statistik, dan explainable AI dapat mendukung sistem peringatan dini berbasis data.

References

[1] S. Boujmiraz, H. Darhmaoui, dan A. D. el maliani, “Predicting Student Performance: A Comprehensive Review of Machine Learning, Deep Learning, and Explainable AI Approaches,” Comput. Educ. Artif. Intell., vol. 10, hal. 100548, 2026, doi: 10.1016/j.caeai.2026.100548.

[2] N. Al-Shanableh et al., “Forecasting Students’ Academic Performance in Educational Data Using Machine Learning Techniques,” Int. J. Inf. Commun. Technol. Educ., vol. 22, no. 1, hal. 1–23, 2026, doi: 10.4018/IJICTE.399756.

[3] H. Zhang, Y. Ren, P. S. Nurius, J. Mankoff, dan A. K. Dey, “Towards Human-Centered Early Prediction Models for Academic Performance in Real-World Contexts,” Proc. ACM Human-Computer Interact., vol. 9, no. 7, hal. 1–41, 2025.

[4] B. B. Alkan, S. Kuzucuk, N. Alkan, dan A. Sinan, “Using Machine Learning to Predict Student Outcomes for Early Intervention and Formative Assessment,” Sci. Rep., vol. 15, hal. 39797, 2025, doi: 10.1038/s41598-025-23409-w.

[5] D. Khairy, N. Alharbi, M. A. Amasha, M. F. Areed, S. Alkhalaf, dan R. A. Abougalala, “Prediction of Student Exam Performance Using Data Mining Classification Algorithms,” Educ. Inf. Technol., vol. 29, no. 16, hal. 21621–21645, 2024, doi: 10.1007/s10639-024-12619-w.

[6] A. G. R. Sandeepa dan S. Mohottala, “Evaluation of Machine Learning Models in Student Academic Performance Prediction,” in 2025 5th International Conference on Advanced Research in Computing (ICARC), 2025, hal. 1–6. doi: 10.1109/ICARC64760.2025.10963104.

[7] H. Gu dan Y. Zhang, “Efficient and Interpretable Machine Learning for Student Academic Outcome Prediction,” Mathematics, vol. 14, no. 4, hal. 626, 2026, doi: 10.3390/math14040626.

[8] C. Wang, J. Chen, Z. Xie, dan J. Zou, “Research on education big data for student’s academic performance analysis based on machine learning,” in Proceedings of the 2024 Guangdong-Hong Kong-Macao Greater Bay Area International Conference on Education Digitalization and Computer Science, 2024, hal. 223–227.

[9] E. Miranda, M. Aryuni, M. I. Rahmawati, S. E. Hiererra, dan A. V. D. Sano, “Machine learning’s model-agnostic interpretability on the prediction of students’ academic performance in video-conference-assisted online learning during the COVID-19 pandemic,” Comput. Educ. Artif. Intell., vol. 7, hal. 100312, 2024, doi: 10.1016/j.caeai.2024.100312.

[10] S. John dan A. Mohamed, “Hybrid-PSP: Hybrid Machine Learning Techniques for Predicting Student Performance,” Procedia Comput. Sci., vol. 282, hal. 2844–2853, 2026, doi: 10.1016/j.procs.2026.06.052.

[11] S. Leelaluk, C. Tang, V. Švábenský, dan A. Shimada, “Knowledge distillation in rnn-attention models for early prediction of student performance,” in Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing, 2025, hal. 64–73.

[12] M. Nagy dan R. Molontay, “Interpretable Dropout Prediction: Towards XAI-Based Personalized Intervention,” Int. J. Artif. Intell. Educ., vol. 34, hal. 274–300, 2024, doi: 10.1007/s40593-023-00331-8.

[13] Y. Chachoui, N. Azizi, R. Hotte, dan T. Bensebaa, “Enhancing algorithmic assessment in education: Equi-fused-data-based SMOTE for balanced learning,” Comput. Educ. Artif. Intell., vol. 6, hal. 100222, 2024, doi: 10.1016/j.caeai.2024.100222.

[14] F. M. Alnassar, T. Blackwell, E. Homayounvala, dan M. Yee-king, “Addressing Class Imbalance in Predicting Student Performance Using SMOTE and GAN Techniques,” Appl. Sci., vol. 16, no. 7, hal. 3274, 2026, doi: 10.3390/app16073274.

[15] W. Hidayatulloh, F. Mahardika, dan D. I. Junaedi, “Explainable artificial intelligence-based model for student academic performance prediction,” J. Inf. Syst. Explor. Res., vol. 4, no. 1, hal. 31–40, 2026, doi: 10.52465/joiser.v4i1.624.

[16] M. M. Islam, F. H. Sojib, M. F. H. Mihad, M. Hasan, dan M. Rahman, “The Integration of Explainable AI in Educational Data Mining for Student Academic Performance Prediction and Support System, ” Telemat. Informatics Reports, vol. 18, hal. 100203, 2025, doi: 10.1016/j.teler.2025.100203.

[17] M. Abdalkareem dan N. Min-Allah, “Explainable Models for Predicting Academic Pathways for High School Students in Saudi Arabia,” IEEE Access, vol. 12, 2024, doi: 10.1109/ACCESS.2024.3369586.

[18] W. Ahmed, M. A. Wani, P. Plawiak, S. Meshoul, A. Mahmoud, dan M. Hammad, “Machine Learning-Based Academic Performance Prediction with Explainability for Enhanced Decision-Making in Educational Institutions,” Sci. Rep., vol. 15, hal. 26879, 2025, doi: 10.1038/s41598-025-12353-4.

[19] K. Hubbard dan S. Amponsah, “Feature Engineering on LMS Data to Optimize Student Performance Prediction,” in New Frontiers in Data Science, H. Han dan J. Stamey, Ed., Cham: Springer Nature Switzerland, 2026, hal. 125–142.

[20] Y. Zhao, H. Wang, dan P. Wang, “Improvement of Applicability in Student Performance Prediction Based on Transfer Learning,” in 2024 5th International Conference on Artificial Intelligence and Electromechanical Automation (AIEA), 2024, hal. 308–314. doi: 10.1109/AIEA62095.2024.10692712.

[21] T. Chen dan C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, in KDD ’16. New York, NY, USA: Association for Computing Machinery, 2016, hal. 785–794. doi: 10.1145/2939672.2939785.

[22] L. Prokhorenkova, G. Gusev, A. Vorobev, A. V. Dorogush, dan A. Gulin, “CatBoost: unbiased boosting with categorical features,” Adv. Neural Inf. Process. Syst., vol. 31, 2018.

[23] M. Mukherjee dan M. Khushi, “SMOTE-ENC: A novel SMOTE-based method to generate synthetic data for nominal and continuous features,” Appl. Syst. Innov., vol. 4, no. 1, hal. 18, 2021.

[24] L. Nguyen, “Student Performance Factors,” https://www.kaggle.com/datasets/lainguyn123/student-performance-factors. Diakses: 20 Juni 2026. [Daring]. Tersedia pada: https://www.kaggle.com/datasets/lainguyn123/student-performance-factors

[25] R. A. Saputri, Asrianda, dan L. Rosnita, “A Random Forest-Based Predictive Model for Student Academic Performance: A Case Study in Indonesian Public High Schools,” J. Appl. Informatics Comput., vol. 9, no. 3, hal. 1042–1049, 2025.

[26] B. Ujkani, D. Minkovska, dan N. Hinov, “Course Success Prediction and Early Identification of At-Risk Students Using Explainable Artificial Intelligence,” Electronics, vol. 13, no. 21, hal. 4157, 2024, doi: 10.3390/electronics13214157.

[27] M. N. Gul, W. Abbasi, M. Z. Babar, A. Aljohani, dan M. Arif, “Data driven decisions in education using a comprehensive machine learning framework for student performance prediction,” Discov. Comput., vol. 28, no. 1, hal. 153, 2025, doi: 10.1007/s10791-025-09585-3.

[28] N. V Chawla, K. W. Bowyer, L. O. Hall, dan W. P. Kegelmeyer, “SMOTE: Synthetic Minority Over-sampling Technique,” J. Artif. Intell. Res., vol. 16, hal. 321–357, 2002, doi: 10.1613/jair.953.

[29] D. Elreedy, A. F. Atiya, dan F. Kamalov, “A theoretical distribution analysis of synthetic minority oversampling technique (SMOTE) for imbalanced learning,” Mach. Learn., vol. 113, no. 7, hal. 4903–4923, 2024, doi: 10.1007/s10994-022-06296-4.

[30] G. S. Hida dan A. C. A. Do Nascimento, “Overview of machine learning in class imbalance scenarios: Trends, challenges, and approaches,” Expert Syst. Appl., vol. 298, hal. 129592, 2026, doi: 10.1016/j.eswa.2025.129592.

[31] L. Breiman, “Random Forests,” Mach. Learn., vol. 45, no. 1, hal. 5–32, 2001, doi: 10.1023/A:1010933404324.

[32] S. M. Lundberg dan S.-I. Lee, “A unified approach to interpreting model predictions,” in Advances in Neural Information Processing Systems, 2017, hal. 4765–4774. [Daring]. Tersedia pada: https://proceedings.neurips.cc/paper/2017/hash/8a20a8621978632d76c43dfd28b67767-Abstract.html

[33] E. Tiukhova et al., “Explainable Learning Analytics: Assessing the stability of student success prediction models by means of explainable AI,” Decis. Support Syst., vol. 182, hal. 114229, 2024, doi: 10.1016/j.dss.2024.114229.

[34] F. T. Johora, M. N. Hasan, A. Rajbongshi, M. Ashrafuzzaman, dan F. Akter, “An explainable AI-based approach for predicting undergraduate students academic performance,” Array, vol. 26, hal. 100384, 2025.

[35] M. Delprato, “Identifying the post-pandemic determinants of low performing students in Latin America through interpretable Machine Learning methods,” Eng. Appl. Artif. Intell., vol. 179, hal. 115154, 2026.