Emotional sentiment analysis of social media content for mental health safety

Abstract

Text sentiment analysis is mostly used for the assessment of the author’s mood depending on the context. The purpose of sentiment analysis is to discover the exactness of the underlying emotion in a given situation. It has been applied to various fields, including stock market predictions, social media data on product evaluations, psychology, the judiciary, forecasting, illness prediction, agriculture, and more. Many researchers have worked on these topics and generated important insights. These outcomes are useful in the field because they (outcomes) help people comprehend the general summary quickly. Additionally, sentiment analysis aids in limiting the harmful effects of some posts on various social media sites such as Facebook and Twitter. For these reasons and more, we are proposing an approach to filter the social media content that could be emotionally harmful to the user, through getting the social networks content; for that, we have used Twitter API to get the user posts, and then, we have used API natural understanding language API tool to extract and classify the emotions of the Twitter content into five basic emotional categories—Joy, sadness, anger, fear, disgust—into an array of emotions; after that, we have defined a perfect emotion array from over 450 words from the English language. The main purpose of this comprehensive research article is to examine the proposed solution that we have conducted to improve the quality of content displayed to users emotionally.