Author : Miss. Shilpa M. Kedari and Prof. Pooja Patil

The development in social network communication prompts the dangerous utilization. An expanding number of social networks mental disorders such as the dependence on the cybernetic relationship, the over-burden of data and the constric- tion of the network, have been noticed recently. Currently, the symptoms of these mental disorders are passively observed, which causes late clinical intervention. In this paper, argue that the mining of online social behavior offers the opportunity to actively identify the mental disorders  at  an  early  stage.  It  is  difficult to detect mental disorders because the mental state cannot be observed directly from the records of online social activities. Our approach, new and innovative for the practice of mental disorders detection it is not based on the self-disclosure of these mental factors through questionnaires psychology. Instead, we propose   a framework of machine learning or the detection of mental disorders in social networks, which exploits the features extracted from social network data to accurately identify potential mental disorders cases. To increase the accuracy of model, we further improve efficiency with performance guarantees. Our framework is evaluated through a user study with no of users of the network. We perform a feature analysis and also apply mental disorders in large-scale data sets and analyze the characteristics of the three types of mental disorder.

Keywords: Online social network (OSN), mental disorder detection, feature extraction, Random forest Classifier