Sentiment Analysis of Mental Health Conversations (Case Study)

The project’s main objective was to create a model to help categorize patients’ messages about mental health by topics.

Three approaches were used to make the model, two of which involved sentiment analysis. The model based on the topic modeling approach was the best. The project also calculated the absolutist index to assist therapists in determining if a patient could have suicidal ideations. Additionally, mental health professionals were rated based on upvotes for their answers to questions. Also, the top 100 most frequent words seen in therapists’ answers to questions on various topics were reported using word clouds. Mental health professionals could use the model as a tool to assist them in providing their patients with the timely help they need.