Depression Detection by Analyzing Social Media Posts of User

Depression is a serious mental health issue for people world-wide irrelevant of their ages, genders and races. In this age of modern communication and technology, people feel more comfortable sharing their thoughts in social networking sites (SNS) almost every day. The objective of this paper is to propose a data-analytic based model to detect depression of any human being. In this proposed model data is collected from the users' posts of two popular social media websites: twitter and facebook. Depression level of a user has been detected based on his posts in social media. The standard method of detecting depression of a person is a fully structured or a semi-structured interview method (SDI) [1]. These methods need a huge amount of data from the person. Micro-blogging sites such as twitter and facebook have become so much popular places to express peoples' activity and thoughts. The data screening from tweets and posts show the manifestation of depressive disorder symptoms of the user. In this research, machine learning is used to process the scrapped data collected from SNS users. Natural Language Processing (NLP), classified using Support Vector Machine (SVM) and Naïve Bayes algorithm to detect depression potentially in a more convenient and efficient way.