Cancer Through the Lens of Social Media: An Examination of Cancer-Related Content on Reddit

Cancer incidence and mortality is increasing globally. The people affected constantly remain in distress the family is under enormous financial and psychological pressure. Social media provides an excellent outlet to vent out all the feelings that develop in one’s mind, which could prove critical for one’s health and survival in cases such as cancer. Reddit is one such social platform with its subreddit dedicated to folks affected by cancer. Subreddit is estimated to be sentiment heavy. We aimed to examine these sentiments and categorize the kind of discussions within the cancer subreddit community, with the intent of categorizing the topic of discussion. Using Python, push shift application programming interface (API) was used nearly ten thousand posts were extracted. After proper data cleaning, we validated the authenticity of data extracted via visualization, then performed Latent Dirichlet Allocation (LDA) to categorize these posts using libraries to separate the posts based on semantics. The topics were categorized in to 3, most topics belonged to category 1 which included emotions, immediate response to diagnosis.