Artificial Intelligence in Detecting Signs of Depression Among Social Networks Users

Mental health has become more than just a concern in recent years, especially depression. Social networks are expanding to such an extent that they have become a platform for assessing mental health based on the analysis of messages and comments from different users. By analyzing social data, we will be able to detect depressive messages. To meet this need for text analysis, it appears that the use of intelligent systems and more particularly Artificial Intelligence is an ideal solution for detecting potential victims. However, the solutions provided are relatively recent and the results vary according to the methodologies employed. Several steps seem also to be essential in building an efficient framework to detect signs of depression in social media users. Thus, in this paper, we present the different phases that a depression detection framework from social data should respect. We then conduct a review comparing existing works in this field. This comparative study highlights the techniques used in every phase. By analyzing research works dealing with this issue, we also aim to understand the way in which Artificial Intelligence (AI) is used, identifying the advantages and limitations of existing works.