EARLY DETECTION OF DEPRESSIVE BEHAVIOR IN ADOLESCENTS THROUGH SOCIAL NETWORKS

Abstract

Due to the anonymity of the Internet and social networking sites, people are stalking in their senses and in the Internet community. For those who do not want to be self-reported, they have to report on cyberbullying in cyberspace and to represent users in cyberspace methods. We focus on the social networking site “VKontakte” and classify users with potential suicidal and suicidal behaviors, with the help of text processing and techniques, based on machine learning.