An Application for Classifying Depression in Tweets
Bernice Yeow Ziwei, Hui Na Chua · 2019
A depressed person feels sad, hopeless on a regular basis, lose interest in activities and experiences physical symptoms such as chronic pain or digestive issues [1]. Each day, millions of people suffered from depression and only a small fraction of them undergo proper treatment. In the past, doctors analyzed depressed people via a face to face session and apply the diagnostic criteria written by a professional psychologist to determine depression. However, past research revealed that most patients would not seek help from doctors at the early stage of depression which results in a declination of their mental health condition. On the other hand, many people are using the social media platform to share their feelings daily. Since then, there have been many studies on using social media to predict mental and physical diseases such as cardiac arrest, zika virus, Muppalla, and prescription drug abuse. However, currently there is a lack of work on an application that leverages social media data for detecting depressive thoughts to determine the depression state of a Twitter user. To achieve this aim, we developed a web application that performs sentiment analysis through a classification function that discerns the proportion of depressive and non-depressive thoughts of an individual. This application further provides individuals to visualize their depression state via a web interface.