Socially Mediated Patient Portal

Abstract

Online Social-Networking Websites (OSNWs) are web-based communities used by individuals to make an online profile, provides a platform through which people share information in very cost effective way and easily expresses their opinions like Facebook, Twitter and MySpace. OSNWs have tremendously altered the way of communication. Face-to-face communications got replaced by posting the status updates, leaving comments and like. These have millions of users, having huge amount of user-generated content (UGC) that can be used in health-related human behaviors study in a cost-effective manner. In worldwide the mental illness is a primary cause of disability. Actually, for the diagnosing of psychological illness, there are no pathological tests; the treatment methodology is highly dependent upon the behavioral actions reported by patient himself and his closed ones. The preciseness and accuracy of information received from patient are subject to his/her non-artificial behaviors. In order to handle the challenge, we found that OSNWs can be used as a screening tool for discovering an affective disorder in individuals. This study investigates how Facebook user’s profile can help in Life care Decision Support System (LCDSS) and how to use Facebook as screening tool to expose the user‘s mental illness from his/her profile. We propose a way to automatically classify Facebook user personally written text: status updates and comments using a Support Vector Machines and other classification algorithms of supervised machine-learning techniques, build a model to set up, train the classifier and identify depressive symptom–related features from user’s profile. The assay of results depicts that statistics of deprived user’s meet DSM-IV criterion for a depression symptoms or Major Depressive Disorder (MDD).