Understanding how daily behavioral patterns contribute to stress is increasingly important in a digitally saturated world. This study investigates the relationship between technology use, lifestyle behaviors, and stress levels using a data-driven machine learning framework. A gradient boosting regressor was trained on a behavioral dataset containing screen time, physical activity, sleep, and dietary patterns, excluding sensitive mental health indicators. The model achieved high predictive accuracy (R² = 0.979), significantly outperforming a baseline. Feature importance and SHAP analyses revealed that social media usage, work-related screen time, and low physical activity were the most impactful predictors of stress, while sleep quality and exercise were associated with lower stress levels. These results suggest that behavioral data alone can provide interpretable and reliable stress assessments, offering practical applications for digital wellness platforms and personalized feedback systems. The findings highlight the importance of monitoring digital habits as a non-clinical approach to mental well-being.
| Mualliflar | Туримов, Д., Сайдирасулов, Н., Киличев, Д. |
|---|---|
| Jurnal | Рақамли технологияларнинг назарий ва амалий масалалари |
| Nashr sanasi | 2025-10-23 |
| Jild | 8 |
| Son | 4 |
| Betlar | 33-40 |
| Til | Ingliz |
| DOI | 10.62132/ijdt.v8i4.300 |
DOI: 10.62132/ijdt.v8i4.300 · Maqolaning asl sahifasi
прогнозирование стресса, использование технологий, машинное обучение, экранное время, поведенческие данные, SHAP, анализ образа жизни, моделирование благополучия, социальные сети, физическая активность, stress prediction, technology use, machine learning, screen time, behavioral data, SHAP, lifestyle analysis, wellness modeling, social media, physical activity
The paper considers the problem of deformation of a thin conductive plate under the action of a non-stationary magnetic field taking into account the Lorentz forces and Joule heat generation. A mathematical model is…
The article considers a numerical method for solving a two-dimensional coupled dynamic thermoplastic boundary value problem based on deformation theory of plasticity. Discrete equations are compiled by the…
High mortality in the acute and post-hospital periods of myocardial infarction underscores the relevance of accurately stratifying patient risk and assessing survival outcomes. This study compares two complementary…
This study introduces a nonlinear mathematical model and a numerical algorithm for simulating the emission and transport of aerosol and dust particles generated by wind erosion of soil in complex orographic terrain. The…
This article develops a model and numerical algorithm with second-order accuracy in time and spatial variables for solving problems of combined heat and moisture transfer during the storage and drying of heterogeneous…
In this research paper, the development and application of parallelization techniques for the efficient solution of systems of logical equations, particularly in the field of medical diagnostics, are described. The…
In this study, a concatenated CNN model for pneumonia detection is proposed, combined with an image enhancement method based on fuzzy logic. The image enhancement process utilizes a novel refined fuzzification…
This review systematizes how machine learning (ML) can be integrated into Lagrangian particle dispersion models (LPDM: FLEXPART, HYSPLIT, STILT, NAME) as a physics-consistent augmentation rather than a replacement of…
The article presents a comprehensive mathematical model of radial filtration in a cylindrical filter, where the fluid flows from the outer layer toward the central cavity. The regions of free flow are described by the…
This paper argues that advances in fuzzy reasoning systems in healthcare are critical for patient data, requiring innovative methodologies that combine the interpretive capabilities of artificial intelligence with…
Рақамли технологияларнинг назарий ва амалий масалалари — barcha maqolalar