Abstract
BACKGROUND: Men who have sex with men (MSM) face distinct external and internal stressors that substantially affect their mental health, leading to higher rates of depression and engagement in risky behaviors. Geosocial networking apps provide new perspectives to assess and promote the mental health of MSM. This study aims to examine the emotions and sentiments of MSM in Shenzhen, China, using data from a Chinese MSM geosocial networking app, and to explore the characteristics, temporal trends, and associated factors of emotions and sentiments. METHODS: This cross-sectional study collected data from the geosocial networking app Blued using a web crawler, focusing on users in Shenzhen who posted more than 15 updates between September and October 2020. Natural language processing techniques, including word segmentation and sentiment analysis, were applied to assess emotional and sentiment patterns. Univariate and multivariate logistic regression analyses were used to identify associated factors with these patterns. RESULTS: The study included 7,537 users with a mean age of 28.7. We found significant emotional fluctuations, particularly negative emotions, during early morning hours. Factors such as body mass index, social media engagement, and the frequency of health-related words were significantly associated with emotional and sentiment trends. Users who posted more frequently and had fewer blacklist entries exhibited more positive emotions. CONCLUSIONS: Social media provides valuable insights into the mental health patterns of MSM. This study highlights the potential for real-time monitoring and targeted interventions tailored to the unique needs of sexual minorities. The findings underscore the importance of culturally relevant and LGBT-affirmative online mental health services.