Komparasi Algoritma Machine Learning Untuk Klasifikasi Risiko Depresi Remaja Berbasis Perilaku Digital Menggunakan SMOTE

Sarwadi Sarwadi; Muhammad Khaibar Putra Adithia

doi:10.70340/jirsi.v5i2.582

Authors

Sarwadi Sarwadi Universitas Nahdaltul Ulama Sumatera Utara
Muhammad Khaibar Putra Adithia Universitas Nahdlatul Ulama Sumatera Utara

DOI:

https://doi.org/10.70340/jirsi.v5i2.582

Keywords:

machine learning, adolescent depression, SMOTE, XGBoost, imbalanced dataset

Abstract

Adolescent depression is an increasingly alarming global mental health issue, particularly amid the rapid adoption of social media platforms. This study compares the performance of four machine learning algorithms Random Forest, XGBoost, Support Vector Machine (SVM), and Logistic Regression, in classifying adolescent depression risk using 12 digital behavioral and lifestyle features. The dataset comprises 1,200 records with severe class imbalance (97.4% vs 2.6%), addressed through Synthetic Minority Over-sampling Technique (SMOTE) with k=5. Eight experiments (four algorithms × two conditions) were conducted and evaluated using Precision, Recall, F1-Macro, AUC-ROC, and 5-fold cross-validation. Results demonstrate that SMOTE consistently enhanced all model performances, with an average F1-Macro improvement of 25.7%. XGBoost achieved perfect F1-Macro and AUC-ROC scores (1.0000) under both conditions, while SMOTE yielded the greatest gains for SVM (+67.9%) and Logistic Regression (+63.7%). Feature importance analysis identified sleep hours, stress level, anxiety level, and daily social media usage as dominant predictors, while academic performance contributed negligibly. These findings underscore the critical importance of imbalanced data handling and provide a methodological framework for developing digital behavior-based early depression detection systems.

Downloads

Download data is not yet available.

References

W. H. Organization, “Depression.” 2021. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/depression

K. K. RI, Riset Kesehatan Dasar (Riskesdas) 2018. Jakarta: Badan Penelitian dan Pengembangan Kesehatan, 2018.

W. A. Social and Hootsuite, “Digital 2023: Indonesia.” 2023. [Online]. Available: https://datareportal.com/reports/digital-2023-indonesia

J. M. Twenge, G. N. Martin, and B. H. Spitzberg, “Trends in U.S. Adolescents’ media use, 1976-2016,” Psychol. Pop. Media Cult., vol. 8, no. 4, pp. 329–345, 2019.

A. Vannucci, K. M. Flannery, and C. M. Ohannessian, “Social media use and anxiety in emerging adults,” J. Affect. Disord., vol. 207, pp. 163–166, 2017.

A. Yadav, S. Bhatt, and A. Kumar, “Machine learning approaches for mental health prediction: A systematic review,” IEEE Access, vol. 10, pp. 41365–41385, 2022.

U. K. J. Dn and M. Rahardi, “Analysis of SMOTE and Random Search on Machine Learning Algorithms for Stroke Disease Diagnosis,” J. Appl. Informatics Comput., vol. 10, no. 1 SE-Articles, pp. 847–855, Feb. 2026, doi: 10.30871/jaic.v10i1.12046.

S. Sarwadi, R. Rosnelly, and B. Triandi, “Feature Selection Analysis for Diagnosing Narcissistic Personality Disorder (NPD) Using Principal Component Analysis and the Naïve Bayes Model,” ZERO J. Sains, Mat. dan Terap., vol. 9, no. 1, p. 14, 2025, doi: 10.30829/zero.v9i1.24086.

A. Surya Firmansyah, A. Aziz, and M. Ahsan, “Optimasi K-Nearest Neighbor Menggunakan Algoritma Smote Untuk Mengatasi Imbalance Class Pada Klasifikasi Analisis Sentimen,” JATI (Jurnal Mhs. Tek. Inform., vol. 7, no. 6, pp. 3341–3347, 2024, doi: 10.36040/jati.v7i6.7257.

G. Almeida and F. Bacao, “Counterfactual synthetic minority oversampling technique: solving healthcare’s imbalanced learning challenge,” Data Sci. Manag., vol. 8, no. 4, pp. 436–446, 2025, doi: https://doi.org/10.1016/j.dsm.2025.01.006.

K. Redwan et al., “Intelligent fault diagnosis in power systems using Random Forest and ensemble learning,” Meas., vol. 10, no. April, p. 100098, 2026, doi: 10.1016/j.meaene.2026.100098.

M. M. Gharagoz, M. Noureldin, and J. Kim, “Explainable machine learning (XML) framework for seismic assessment of structures using Extreme Gradient Boosting (XGBoost),” Eng. Struct., vol. 327, no. January, p. 119621, 2025, doi: 10.1016/j.engstruct.2025.119621.

G. O. Anyanwu, C. I. Nwakanma, J.-M. Lee, and D.-S. Kim, “RBF-SVM kernel-based model for detecting DDoS attacks in SDN integrated vehicular network,” Ad Hoc Networks, vol. 140, p. 103026, 2023, doi: https://doi.org/10.1016/j.adhoc.2022.103026.

L. Sammut, P. Bezzina, V. Gibbs, Y. Muscat-Baron, A. Agius-Camenzuli, and J. Calleja-Agius, “Predicting first-trimester pregnancy outcome in threatened miscarriage: A comparison of a multivariate logistic regression and machine learning models,” Radiography, vol. 31, no. 6, p. 103159, 2025, doi: 10.1016/j.radi.2025.103159.

J. Y. Verbakel et al., “ROC curves for clinical prediction models part 1. ROC plots showed no added value above the AUC when evaluating the performance of clinical prediction models,” J. Clin. Epidemiol., vol. 126, pp. 207–216, 2020, doi: https://doi.org/10.1016/j.jclinepi.2020.01.028.

Z. Han, L. Wang, H. Zhu, Y. Tu, P. He, and B. Li, “Uncovering the effects and mechanisms of tea and its components on depression, anxiety, and sleep disorders: A comprehensive review,” Food Res. Int., vol. 197, p. 115191, 2024, doi: https://doi.org/10.1016/j.foodres.2024.115191.

T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system BT - Proc. 22nd ACM SIGKDD International Conference on KDD,” 2016, pp. 785–794.

G. Schrijver, D. K. Sarmah, and M. El-hajj, “Automobile insurance fraud detection using data mining: A systematic literature review,” Intell. Syst. with Appl., vol. 21, no. November 2023, p. 200340, 2024, doi: 10.1016/j.iswa.2024.200340.

I. Abunadi, “Machine learning techniques for analysing cardiotocography signals for early detection of fetal anomalies based on feature engineering methods,” Mach. Learn. with Appl., vol. 22, no. October, p. 100766, 2025, doi: 10.1016/j.mlwa.2025.100766.

H. Zhou, J. Tong, Y. Liu, K. Zheng, and C. Cao, “An oversampling FCM-KSMOTE algorithm for imbalanced data classification,” J. King Saud Univ. - Comput. Inf. Sci., vol. 36, no. 10, p. 102248, 2024, doi: 10.1016/j.jksuci.2024.102248.

A. G. Reece and C. M. Danforth, “Instagram photos reveal predictive markers of depression,” EPJ Data Sci., vol. 6, no. 15, pp. 1–12, 2017.