Evaluating Turkish society's approaches to hemorrhoid treatment using text mining and sentiment analysis methods.

Background: Hemorrhoids is a rectal disease that has adverse effects on life quality. Because the disease occurs in a discreet area, the rectal, patients often hesitate to consult a physician out of embarrassment. Thus, they often turn to the internet to find support and solutions. The ultimate objective of this study is to determine the societal attitudes, behaviors, and perceptions towards hemorrhoids in Turkey. To achieve this, a comparative sentiment analysis of social media comments was performed for the purpose of getting an idea about different treatment methods' effectiveness and satisfaction rate. Methods: The data set used includes user comments on YouTube videos in Turkish on the topic, which are open to public access. In this regard, totally 15.382 comments under 116 videos were analyzed. Sentiment analysis using a dictionary-based model was done, followed by the process of obtaining a sentiment score for each comment via the VADER model. The comments were divided into four groups (treatment methods) using the LDA (Latent Dirichlet Allocation) method based on the words related to their subject matter: surgical, pharmaceutical, herbal, and traditional. Results: In terms of sentiment scores regarding treatment methods (VADER), the differences among YouTube users were found to be statistically significant (KW-H = 43.873, Conclusion: Turkish societal attitudes, feelings, and treatment methods preferences toward hemorrhoids were determined through sentiment analysis and text mining methods. Consequently, data obtained from social media tools provides valuable insight into societal health-related tendencies and behaviors.