<p>Online social media platforms have become extremely popular for studying mental health disorders. This study examines conversations about mental health on social media sites. The objective is to examine user-generated content from different subreddits pertaining to mental health in order to find recurring themes and patterns of behavior linked to illnesses on subreddits obtained on schizophrenia, bipolar disorder, anxiety, depression, autism, borderline personality disorder, and mental health from Reddit. The study extracts topics using Latent Dirichlet Allocation (LDA) and BERTopic assesses them using a variety of coherence metrics and perplexity. The findings demonstrate how social media data can provide complex insights into mental health, supporting early diagnosis and the creation of focused interventions. This study highlights how crucial it is to use social media to learn about patients’ viewpoints in addition to conventional clinical methods.</p>