Perbandingan Model Regresi Machine Learning untuk Prediksi Skor Tingkat Stres Berdasarkan Pola Screen Time Tahun 2025
DOI:
https://doi.org/10.52436/1.jpti.1537Kata Kunci:
Kesehatan Mental, Machine Learning, Prediksi Stres, Screen Time, XGBoostAbstrak
Transformasi digital yang masif pada era modern telah mendorong peningkatan signifikan dalam durasi paparan layar (screen time), yang diidentifikasi sebagai salah satu faktor risiko utama terhadap kesehatan mental, khususnya peningkatan prevalensi stres psikologis. Metode diagnosis konvensional yang mengandalkan instrumen kuesioner mandiri dinilai kurang optimal karena rentan terhadap bias pelaporan dan bersifat subjektif. Penelitian ini bertujuan untuk membandingkan performa tiga algoritma machine learning, yaitu Random Forest, Support Vector Regression (SVR), dan XGBoost Regression, dalam memprediksi skor tingkat stres secara kontinu (skala 0–10) berdasarkan pola penggunaan perangkat digital. Tahapan penelitian meliputi akuisisi dataset "Screentime vs Mental Wellness Survey 2025" dari repositori publik, pra-pemrosesan data melalui imputasi statistik, normalisasi Min-Max Scaling, dan One-Hot Encoding, dilanjutkan dengan pembangunan model menggunakan evaluasi 10-fold cross-validation serta interpretasi model berbasis metode SHAP. Hasil evaluasi pada data uji menunjukkan bahwa XGBoost merupakan model dengan performa terbaik, mencapai nilai Mean Absolute Error (MAE) terendah sebesar 0,6502, Root Mean Squared Error (RMSE) sebesar 0,8253, dan koefisien determinasi (R²) sebesar 0,8367. Temuan ini mengindikasikan bahwa model mampu menjelaskan lebih dari 83% variasi tingkat stres pada data yang belum pernah dilatih sebelumnya. Analisis feature importance mengungkapkan bahwa indeks kesejahteraan mental dan produktivitas merupakan prediktor paling dominan, sedangkan durasi screen time berkontribusi relatif kecil, yang menunjukkan bahwa faktor psikologis internal lebih berpengaruh terhadap stres dibandingkan intensitas interaksi digital semata. Penelitian ini menyimpulkan bahwa pendekatan ensemble learning, khususnya XGBoost, efektif dalam memodelkan fenomena stres yang bersifat kompleks dan multidimensional sebagai dasar pengambilan keputusan klinis berbasis data.
Unduhan
Referensi
C. Pieh et al., “Smartphone screen time reduction improves mental health: a randomized controlled trial,” BMC Med., vol. 23, no. 1, 2025, doi: 10.1186/s12916-025-03944-z.
C. Zhu, S. Li, and L. Zhang, “The impact of smartphone addiction on mental health and its relationship with life satisfaction in the post-COVID-19 era,” Front. Psychiatry, vol. 16, no. March, pp. 1–10, 2025, doi: 10.3389/fpsyt.2025.1542040.
M. Abd Al-Alim, R. Mubarak, N. M. Salem, and I. Sadek, “A machine-learning approach for stress detection using wearable sensors in free-living environments,” Comput. Biol. Med., vol. 179, 2024, doi: 10.1016/j.compbiomed.2024.108918.
O. Higgins and R. L. Wilson, “Integrating Artificial Intelligence (AI) With Workforce Solutions for Sustainable Care: A Follow Up to Artificial Intelligence and Machine Learning (ML) Based Decision Support Systems in Mental Health,” Int. J. Ment. Health Nurs., vol. 34, no. 2, pp. 1–10, 2025, doi: 10.1111/inm.70019.
S. K. Sharma, A. I. Alutaibi, A. R. Khan, G. G. Tejani, F. Ahmad, and S. J. Mousavirad, “Early detection of mental health disorders using machine learning models using behavioral and voice data analysis,” Sci. Rep., vol. 15, no. 1, pp. 1–19, 2025, doi: 10.1038/s41598-025-00386-8.
F. D. E. Baran and M. Cetin, “AI-driven early diagnosis of specific mental disorders: a comprehensive study,” Cogn. Neurodyn., vol. 19, no. 1, 2025, doi: 10.1007/s11571-025-10253-x.
M. Payne et al., “Enhancing Mental Health Decision-Making with Artificial Intelligence / Machine Learning?: A Prescriptive Analytics Approach for Customised Outcomes,” medRxiv, 2025, [Online]. Available: https://www.medrxiv.org/content/10.1101/2025.06.16.25329741.abstract%0Ahttps://www.medrxiv.org/content/medrxiv/early/2025/06/17/2025.06.16.25329741.full.pdf
L. Zhang, S. Zhao, Z. Yang, H. Zheng, and M. Lei, “An artificial intelligence tool to assess the risk of severe mental distress among college students in terms of demographics, eating habits, lifestyles, and sport habits: an externally validated study using machine learning,” BMC Psychiatry, vol. 24, no. 1, 2024, doi: 10.1186/s12888-024-06017-2.
Y. Zhang et al., “Employing Machine Learning and Deep Learning Models for Mental Illness Detection,” Computation, vol. 13, no. 8, pp. 1–14, 2025, doi: 10.3390/computation13080186.
L. H. Milasan and D. Scott-Purdy, “The Future of Artificial Intelligence in Mental Health Nursing Practice: An Integrative Review,” Int. J. Ment. Health Nurs., vol. 34, no. 1, pp. 1–19, 2025, doi: 10.1111/inm.70003.
M. Al Fachrozi and K. D. Tania, “Comparison of Naive Bayes, SVM, K-NN, Decision Tree, and Random Forest in Sentiment Analysis Based on SeaBank Application Aspects,” vol. XII, no. 1, 2025.
D. Amanda Ardhani and K. D. Tania, “Knowledge Discovery on E-Commerce Customer Churn Using Interpretable Machine Learning: A Comparative Study of SHAP-Based Classifiers,” J. Appl. Informatics Comput., vol. 9, no. 5, pp. 2695–2702, 2025, doi: 10.30871/jaic.v9i5.10811.
D. N. Fahria, K. D. Tania, and R. D. Kurnia, “Analisis komparatif algoritma random forest, xgboost, dan catboost untuk klasifikasi tingkat stres pengguna media sosial 1),” vol. 11, no. 1, pp. 1843–1853, 2026.
C. El Morr et al., “Predictive Machine Learning Models for Assessing Lebanese University Students’ Depression, Anxiety, and Stress During COVID-19,” J. Prim. Care Community Heal., vol. 15, 2024, doi: 10.1177/21501319241235588.
B. Carter, M. Payne, P. Rees, S. Y. Sohn, J. Brown, and N. J. Kalk, “A multi-school study in England, to assess problematic smartphone usage and anxiety and depression,” Acta Paediatr. Int. J. Paediatr., vol. 113, no. 10, pp. 2240–2248, 2024, doi: 10.1111/apa.17317.
