Clustering Help-Seeking Behaviors In Lgbt Online Communities: A Prospective Trial

Online Lesbian, Gay, Bisexual, and Transgender (LGBT) support communities have emerged as a major social media platform for sexual and gender minorities (SGM). These communities play a crucial role in providing LGBT individuals a private and safe space for networking because LGBT individuals are more likely to experience social isolation and family rejection. However, the emergence of these online communities introduced new public health concerns and challenges. Since LGBT individuals are vulnerable to mental illness and risk of suicide as compared to the heterosexual population, crisis prevention and intervention are important. Nevertheless, such a protection mechanism has not yet become a serious consideration when it comes to the design of LGBT online support communities partially because of the difficulties of identifying at-risk users effectively and timely. This pilot study aims to explore the potential of identifying LGBT user discussions related to help-seeking through natural language processing and topic model. The findings suggest the feasibility of the proposed approach by identifying topics and representative forum discussions that contain help-seeking information. This study provides important data to suggest the future direction of improving data analytics and computer-aided modules for LGBT online communities with the goal of enhancing crisis suicide prevention and intervention.