Machine learning prediction models for the popularization and dissemination of medical science popularization videos DOI Creative Commons

Nuo Cheng,

Xiuling Wang, Yang Mu

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 19, 2024

Abstract Objective To summarize the current shooting trends of this type video, discuss effect non-medical factors on spread videos, and develop prediction models using machine learning (ML) algorithms. Methods We searched filtered medical science popularization videos TikTok, then labeled features as variables record number “Thumb-Up”, “Comment”, “Share” “Collection” outcome indicators. A total 286 samples 34 were included in construction ML model, 13 algorithms employed with area under curve (AUC) for performance assessment a ten-fold cross-validation accuracy testing. Results In quantitative analysis 4 indicators, we identified significant disparities among different videos. Subsequently, five best-performing ultimately confirmed to predict reasons differences: “Thumb-Up” RF Model (AUC = 0.7331), 0.7439), 0.7077), “Comment” 0.7960), BNB 0.7844). By models, video duration, title description length, location emerged body language most crucial parameters across all models. Conclusion demonstrated superior predicting influence The weight these will provide valuable guidance preparation. This study contributes dissemination acceptance by public, thereby promoting health education enhancing public awareness competence healthcare.

Language: Английский

Digital gratification: short video consumption and mental health in rural China DOI Creative Commons
Chen Zhang,

Bochen Zhu

Frontiers in Public Health, Journal Year: 2025, Volume and Issue: 13

Published: April 22, 2025

Background In recent years, short videos have become increasingly popular in rural China, yet their impact on mental health remains underexplored. While prior studies debated the psychological effects of social media, little is known about how short-form video consumption affects populations. Objective This study investigates causal relationship between and among residents China. Methods We use longitudinal data from China Family Panel Studies apply a Difference-in-Differences strategy to estimate frequent usage health. To address self-selection staggered treatment timing, we employ Propensity Score Matching heterogeneity-robust difference-in-differences estimators. Robustness checks include placebo tests an event analysis. Results find that appears improve residents. The effect immediate significant only first year exposure, but fades subsequent periods. Mechanism analysis suggests improvements are driven by enhanced entertainment information access rather than increased interaction. more pronounced economically underdeveloped less pandemic-affected regions, not evident urban Conclusion Short provide short-term benefits for Chinese enriching leisure access, especially developed areas. However, positive transient cannot offset pandemic-related stress. Policy efforts should aim balance digital with potential risks such as addiction overload.

Language: Английский

Citations

0

Stress Coping Strategies on Short Video Social Media [Letter] DOI Creative Commons
M. Zaenul Muttaqin

Psychology Research and Behavior Management, Journal Year: 2024, Volume and Issue: Volume 17, P. 217 - 218

Published: Jan. 1, 2024

Exploring Stress Coping Strategies on Short Video Social Media During the COVID-19 Pandemic."This research is superior in several ways: 1) this somewhat different from general research, which emphasizes negative impact of social media psychology, while specifically analyzes implications short videos with positive figurations personal well-being; 2) constructs a stress management model that correlated behaviour platform users during lockdown policy; 3) focusing video platforms and communication between problem-focus coping emotion-focus features, thereby offering comprehensive understanding viewing lockdown. 1owever, notes need to be taken into consideration by authors: although situation-strategy plays an important role person-situation interactionist theory, it add Problem-Focused (PFC) factors as emphasized Lazarus Folkman such health energy or beliefs; claimed have implications, results addiction are difficult simplify; data collection via media, so information provided may limited not in-depth.To gain deeper understanding, future can embrace note through Research includes many influence coping; comparative impacts stress; Qualitative design approaches case studies virtual ethnography, obtain more in-depth regarding coping. 2

Language: Английский

Citations

1

Machine learning prediction models for the popularization and dissemination of medical science popularization videos DOI Creative Commons

Nuo Cheng,

Xiuling Wang, Yang Mu

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 19, 2024

Abstract Objective To summarize the current shooting trends of this type video, discuss effect non-medical factors on spread videos, and develop prediction models using machine learning (ML) algorithms. Methods We searched filtered medical science popularization videos TikTok, then labeled features as variables record number “Thumb-Up”, “Comment”, “Share” “Collection” outcome indicators. A total 286 samples 34 were included in construction ML model, 13 algorithms employed with area under curve (AUC) for performance assessment a ten-fold cross-validation accuracy testing. Results In quantitative analysis 4 indicators, we identified significant disparities among different videos. Subsequently, five best-performing ultimately confirmed to predict reasons differences: “Thumb-Up” RF Model (AUC = 0.7331), 0.7439), 0.7077), “Comment” 0.7960), BNB 0.7844). By models, video duration, title description length, location emerged body language most crucial parameters across all models. Conclusion demonstrated superior predicting influence The weight these will provide valuable guidance preparation. This study contributes dissemination acceptance by public, thereby promoting health education enhancing public awareness competence healthcare.

Language: Английский

Citations

0