Depression Detection and Prevention System by Analysing Tweets
6 Pages Posted: 25 Mar 2019
Date Written: March 23, 2019
Abstract
Social media platforms like Twitter which is a microblogging tool enable its users to express their feelings, emotions and opinions through short text messages. Detecting the emotions in a text can help one identify anxiety and depression of an individual. Depression is a mental health problem which can happen to anyone, at any age. There is a lack of systematic and efficient methods to identify the psychological state of an individual. With more than 58 millions tweets generated daily, Twitter can be used in order to detect the sign of depression in a faster way. Recent studies have demonstrated that Twitter can be used to prevent one from taking an extreme step. Our Proposed depression detection and prevention system can detect any depression related words or phrases from Tweets and also classify the type of depression, if detected. This system is proposed in order to diagnose depression and prevent it. Proposed system is using Support vector machine and Naïve Bayes classifier. This hybrid approach works well not only with shorter snippets but also with longer snippets.
Keywords: Natural Language Processing, Machine learning, Twitter Analysis
Suggested Citation: Suggested Citation