Comparing TabNet and CatBoost Models for Explainable Student Depression Prediction
-
Triana Dewi Salma
*
Universitas LIA, Jakarta,
Indonesia
- Rizqi Darmawan Universitas LIA, Jakarta, Indonesia
- Fauzan Natsir Universitas Indraprasta PGRI, Jakarta, Indonesia
- Esa Kurniawan Universitas LIA, Jakarta, Indonesia (*) Corresponding Author
Abstract
Early identification of depression risk among students is increasingly important for educational institutions because mental health problems may affect academic engagement, well-being, and learning continuity. This study aims to compare TabNet and CatBoost in predicting student depression risk and to examine their explainability in identifying influential predictors from structured tabular data. The dataset used in this study consists of student-related variables covering personal characteristics, academic conditions, psychological indicators, and lifestyle factors. The experimental procedure included data cleaning, missing value treatment, categorical feature transformation, feature scaling, model training, testing, and interpretation. Model performance was assessed using accuracy, precision, recall, F1-score, and ROC-AUC. Meanwhile, model explainability was examined through attention-based feature importance for TabNet and SHAP-based interpretation for CatBoost. The experimental results indicate that CatBoost produced better overall classification performance, achieving 84.54% accuracy compared with 83.44% for TabNet. CatBoost also obtained higher precision, F1-score, and ROC-AUC values. In contrast, TabNet showed slightly better recall, suggesting stronger sensitivity in detecting students classified as at risk. The interpretation results show that suicidal thoughts, financial stress, academic pressure, sleep duration, study satisfaction, dietary habits, and workload-related variables were consistently relevant to the prediction process. These findings indicate that model selection for student depression prediction should consider not only accuracy, but also sensitivity and interpretability.
Downloads
References
Y. Deng et al., “Family and Academic Stress and Their Impact on Students’ Depression Level and Academic Performance,” Frontiers in Psychiatry, vol. 13, p., 2022, doi: 10.3389/fpsyt.2022.869337.
G. Gómez-García, M. Ramos-Navas-Parejo, J. C. D. L. Cruz-Campos, and C. Rodríguez-Jiménez, “Impact of COVID-19 on University Students: An Analysis of Its Influence on Psychological and Academic Factors,” International Journal of Environmental Research and Public Health, vol. 19, p., 2022, doi: 10.3390/ijerph191610433.
G. Ahmed, A. Negash, H. Kerebih, D. Alemu, and Y. Tesfaye, “Prevalence and associated factors of depression among Jimma University students. A cross-sectional study,” International Journal of Mental Health Systems, vol. 14, p., 2020, doi: 10.1186/s13033-020-00384-5.
C. Son, S. Hegde, A. Smith, X. Wang, and F. Sasangohar, “Effects of COVID-19 on College Students’ Mental Health in the United States: Interview Survey Study,” Journal of Medical Internet Research, vol. 22, p., 2020, doi: 10.2196/21279.
X. Wang, S. Hegde, C. Son, B. Keller, A. Smith, and F. Sasangohar, “Investigating Mental Health of US College Students During the COVID-19 Pandemic: Cross-Sectional Survey Study,” Journal of Medical Internet Research, vol. 22, p., 2020, doi: 10.2196/22817.
B. Choi, G. Shim, B. Jeong, and S.-C. Jo, “Data-driven analysis using multiple self-report questionnaires to identify college students at high risk of depressive disorder,” Scientific Reports, vol. 10, p., 2020, doi: 10.1038/s41598-020-64709-7.
C. Hobbs, G. Lewis, C. Dowrick, D. Kounali, T. Peters, and G. Lewis, “Comparison between self-administered depression questionnaires and patients’ own views of changes in their mood: a prospective cohort study in primary care,” Psychological Medicine, vol. 51, pp. 853–860, 2020, doi: 10.1017/S0033291719003878.
C. Sun, S. Li, D. Cao, F.-Y. Wang, and A. Khajepour, “Tabular Learning-Based Traffic Event Prediction for Intelligent Social Transportation System,” IEEE Transactions on Computational Social Systems, vol. 10, pp. 1199–1210, 2023, doi: 10.1109/TCSS.2022.3170934.
H. Nguyen and H. Byeon, “Predicting Depression during the COVID-19 Pandemic Using Interpretable TabNet: A Case Study in South Korea,” Mathematics, p., 2023, doi: 10.3390/math11143145.
C. Zhang, X. Chen, S. Wang, J. Hu, C. Wang, and X. Liu, “Using CatBoost algorithm to identify middle-aged and elderly depression, national health and nutrition examination survey 2011–2018,” Psychiatry Research, vol. 306, p. 114261, 2021, doi: https://doi.org/10.1016/j.psychres.2021.114261.
S. Arik and T. Pfister, “TabNet: Attentive Interpretable Tabular Learning,” ArXiv, vol. abs/1908.07442, p., 2019, doi: 10.1609/aaai.v35i8.16826.
D. Enkhbayar et al., “Explainable Artificial Intelligence Models for Predicting Depression Based on Polysomnographic Phenotypes,” Bioengineering, vol. 12, no. 2, p. 186, Feb. 2025, doi: 10.3390/bioengineering12020186.
L. Yang et al., “Application of machine learning in depression risk prediction for connective tissue diseases,” Sci Rep, vol. 15, no. 1, p. 1706, Jan. 2025, doi: 10.1038/s41598-025-85890-7.
T. D. Salma, G. A. P. Saptawati, and Y. Rusmawati, “Text Classification Using XLNet with Infomap Automatic Labeling Process,” in 2021 8th International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA), 2021, pp. 1–6. doi: 10.1109/ICAICTA53211.2021.9640255.
T. D. Salma, M. F. Kurniawan, R. Darmawan, and A. Basri, “Analisis Sentimen Berbasis Transformer: Persepsi Publik terhadap Nusantara pada Perayaan Kemerdekaan Indonesia yang Pertama,” J. JTIK (Jurnal Teknol. Inf. dan Komunikasi), vol. 9, no. 2, pp. 757–764, 2025.
J. Si, W. Y. Cheng, M. Cooper, and R. Krishnan, “InterpreTabNet: Distilling Predictive Signals from Tabular Data by Salient Feature Interpretation,” ArXiv, vol. abs/2406.00426, p., 2024, doi: 10.48550/arXiv.2406.00426.
J. Hancock and T. Khoshgoftaar, “CatBoost for big data: an interdisciplinary review,” Journal of Big Data, p., 2020, doi: 10.1186/s40537-020-00369-8.
L. Prokhorenkova, G. Gusev, A. Vorobev, A. V. Dorogush, and A. Gulin, “CatBoost: unbiased boosting with categorical features,” Jun. 2017, [Online]. Available: http://arxiv.org/abs/1706.09516
S. Lundberg and S.-I. Lee, “A Unified Approach to Interpreting Model Predictions,” May 2017, [Online]. Available: http://arxiv.org/abs/1705.07874
S. Somvanshi, S. Das, S. A. Javed, G. Antariksa, and A. Hossain, “A Survey on Deep Tabular Learning,” p., 2024.
“Student Depression Dataset.” Accessed: May 15, 2026. [Online]. Available: https://www.kaggle.com/datasets/hopesb/student-depression-dataset
Bila bermanfaat silahkan share artikel ini
Berikan Komentar Anda terhadap artikel Comparing TabNet and CatBoost Models for Explainable Student Depression Prediction
Pages: 268-277
Copyright (c) 2026 Triana Dewi Salma, Rizqi Darmawan, Fauzan Natsir, Esa Kurniawan
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (Refer to The Effect of Open Access).
