Predicting Personal Traits from Vkontakte Images

— Psychometric analysis of information from the Internet is one of the fastest growing trends in modern research. Using social networks, one can identify psychological characteristics and mental disorders. In this paper, we solve the problem of identifying personal traits, that is, predictors of depression among users of the VKontakte social network by analyzing the images they publish. We describe our methods and approaches for solving this problem and present the results of experimental verification on Vkontakte data. Our study shows that one can use object detection methods to create effective features for predicting personality traits.