Their Post Tell theTruth: Detecting Social Media Users Mental Health Issues with Sentiment Analysis

Abstract

Mental health disorders remain a problem that always appears throughout the ages because its originators are related to everyday social phenomena that are always changing. One of the serious obstacles is cultural factors that view people with mental health disorders as people who cannot function fully, need to be avoided, have problems, and are given a negative social stigma. On the other hand, people with mental health disorders need a comfortable space where they can express their emotions and thoughts. Social media such as Twitter is one of the potential cathartic media for people with mental health disorders. This study aims to identify mental health disorders through words or tweet narration with the keywords “emotions”, “hallucinations”, “panic”, “mental illness”, “stress”, and “fear”. 5537 clean tweet data were collected from the Indonesian population containing these keywords using rapidminer sentiment analysis which were categorized into three categories; Positive, Negative, and Neutral. The results of the analysis are strengthened by searching randomly sampled text tweets. As a result, it is proven that social media Twitter is effective in identifying symptoms of mental health disorders, and Twitter is considered a safe and comfortable cathartic medium for people with mental health disorders.