This paper explores the potential of using social media data to detect and predict major depressive disorder (MDD). The authors use crowdsourcing to collect data from Twitter users who report being diagnosed with clinical depression, based on a standard psychometric instrument. They analyze behavioral attributes such as social engagement, emotion, language, and linguistic styles, as well as ego networks and mentions of antidepressant medications. The study finds that social media contains useful signals for characterizing the onset of depression, including decreased social activity, increased negative emotion, high self-attentional focus, and heightened concerns about relationships and medication. The authors build a statistical classifier that can predict the risk of depression before the reported onset, achieving an accuracy of 70% and precision of 0.74. The study highlights the potential of social media as a tool for identifying the onset of major depression, which could be used by healthcare agencies or individuals to take proactive steps regarding their mental health. The research also demonstrates that social media data can provide valuable insights into the behavioral and emotional patterns of individuals with depression, including their use of depression-related language and mentions of antidepressant medications. The findings suggest that social media activity can serve as a useful signal for predicting depression, and that the analysis of social media data can contribute to the development of tools for early detection and intervention. The study also discusses the implications of using social media data for mental health monitoring, including privacy considerations and the potential for developing scalable methods for automated public health tracking.This paper explores the potential of using social media data to detect and predict major depressive disorder (MDD). The authors use crowdsourcing to collect data from Twitter users who report being diagnosed with clinical depression, based on a standard psychometric instrument. They analyze behavioral attributes such as social engagement, emotion, language, and linguistic styles, as well as ego networks and mentions of antidepressant medications. The study finds that social media contains useful signals for characterizing the onset of depression, including decreased social activity, increased negative emotion, high self-attentional focus, and heightened concerns about relationships and medication. The authors build a statistical classifier that can predict the risk of depression before the reported onset, achieving an accuracy of 70% and precision of 0.74. The study highlights the potential of social media as a tool for identifying the onset of major depression, which could be used by healthcare agencies or individuals to take proactive steps regarding their mental health. The research also demonstrates that social media data can provide valuable insights into the behavioral and emotional patterns of individuals with depression, including their use of depression-related language and mentions of antidepressant medications. The findings suggest that social media activity can serve as a useful signal for predicting depression, and that the analysis of social media data can contribute to the development of tools for early detection and intervention. The study also discusses the implications of using social media data for mental health monitoring, including privacy considerations and the potential for developing scalable methods for automated public health tracking.