Introduction

Social media platforms significantly influence public perception and individual behaviour, particularly regarding aesthetic enhancements. Instagram, TikTok, Douyin, Kuaishou, X, Sina Weibo, and VK showcase content related to various aesthetic procedures, shaping societal norms around beauty and self-image. Despite the prevalence of this content, understanding its psychological impact and societal attitudes remains underexplore.

Methods

We analysed over 14.9 million social media posts related to aesthetic enhancements from seven platforms, collected between January 2019 and January 2024. Data collection utilized platform-specific APIs and web scraping, focusing on relevant keywords and hashtags. Posts were cleaned, normalized, and translated. Sentiment analysis used VADER and machine learning models (logistic regression, SVM, random forest, and BERT). Psychological factors were identified using latent Dirichlet allocation (LDA) and Bayesian modelling.

Results

Initial VADER analysis categorized sentiments as 45% positive, 30% neutral, and 25% negative, with an 85% accuracy. The BERT-based model improved accuracy to 92%. Positive sentiments peaked during Summer, neutral sentiments were highest in April, and negative sentiments remained stable. Psychological analysis revealed a strong positive correlation between self-esteem and positive sentiments, while societal pressure was negatively correlated. Younger users and females exhibited significant variations in sentiment and psychological factors.

Conclusion

This study provides a comprehensive analysis of aesthetic enhancement discourse on social media, revealing seasonal and demographic sentiment variations and profound psychological impacts. These insights are crucial for practitioners in the aesthetic industry and mental health professionals to tailor strategies and support mechanisms. The study emphasizes the need for responsible messaging and realistic beauty representations to mitigate negative psychological effects.

Level of evidence iv

This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .