Author : Miss. Prajkta Lahu Jagdale and Prof.Dr. Archana Lomte

The advancement in social network communication prompts the dangerous usage. An extending number of social networks mental issue (SNMI), such as the dependence on the cybernetic relationship, the over-burden of data and the constriction of the network, have been noticed recently. As of now, the side effects of these psychological issue are latently watched, which causes late clinical intercession. In this paper, contend that the mining of online social conduct offers the chance to effectively recognize the SNMI at a beginning time. It is hard to identify SNMI in light of the fact that the psychological state can’t be watched straightforwardly from the records of online social activities. Our methodology, new and imaginative for the act of SNMI location, it did not depend on the self-divulgence of these psychological factors through surveys brain science. Rather, we propose a system of reinforement learning, or the detection of mental disorders in social networks (SNMI), which exploits the features extracted from social network data to accurately identify potential SNMI cases. We also use multiple sources learning in SNMI. and proposing a new SNMI-weka tool to improve accuracy. To increase the scalability of STM, further improve efficiency with performance guarantees. Our system is assessed is evaluated through a user study with no of users of the network. This system perform a feature analysis and also apply SNMI in large-scale data sets and analyze the characteristics of the three types of mental disorder.

Keywords: social network, mental disorder detection, feature extraction, Q-Learning classifier