Detecting early signs of depression using user-generated data from social media

Abstract

Social media has now become the most popular medium for self-expression and community building. More and more people are turning to it to express, write, vent or just be while being comfortable doing so because it’s an accepted social norm. Using a dataset extracted from Twitter of depressed and non-depressed users, this project aims at studying the characteristic features that make suitable early indicators of depression in Twitter-Users, hence early detection of depression. The project has to put to application Natural Language Processing and Machine Learning algorithms. In its first phase, the project at the moment uses text mining, assigning moods to different tweets, predicting change in moods, running experiments on non-voluntary data.