Understanding Mental Health Concerns: A Topic Modeling Approach to User Inquiries on a Counseling Platform

This study explores mental health concerns on the Iranian online platform Ninisite by applying topic modeling to user-generated content. It addresses challenges such as delayed responses and unqualified experts by analyzing user interactions through Latent Dirichlet Allocation (LDA). The goal is to uncover prevalent mental health issues, improve service delivery, and enhance user experience. Data was collected from NiniSite using web scraping, targeting ۵۰ psychological hashtags and resulting in a dataset of ۳۲۴۲ inquiries. The LDA model identified five distinct topics: relationship difficulties, family and parenting struggles, child-related concerns, sexual health and parental issues, and psychological and behavioral challenges. A coherence score of ۰.۴۴۶۹ indicated reasonably coherent and interpretable topics. The findings highlight the interconnectedness of family dynamics, relationship struggles, and mental health, emphasizing the need for culturally sensitive, open dialogue, and tailored intervention strategies to support Iranian women's mental health.

کلیدواژه ها:

Mental health ، online platforms ، topic modeling ، user inquiries ، Ninisite