Social media data is being increasingly used to computationally learn about and infer behaviors and underlying psychologies of people. Adopting a social ecological perspective, this talk highlights a series of approaches that allow the gleaning of potentially useful mental health biomarkers by harnessing. At the same time, the talk discusses a critical analytic study surrounding the pitfalls of social media biomarkers of mental health -- in particular, the theoretical, domain, and psychometric validity of this novel information source as well as its underlying biases, when appropriated to augment conventionally gathered biomarkers, such as those gathered via surveys and verbal self-reports.
Then, to overcome these pitfalls, this talk presents results from a case study, where computational algorithms to glean mental health biomarkers from social media were developed in a context-sensitive and human-centered way, in collaboration with domain experts and stakeholders. Specifically, this study, a collaboration with a health provider, reveals the ability and implications of using social media data of consented schizophrenia patients to forecast relapse and support clinical decision-making. The talk concludes with provocations of the path forward, emphasizing the need for a collaborative, multidisciplinary research agenda while realizing the potential of social media derived biomarkers in mental health -- one that incorporates methodological rigor, ethics, and accountability, all at once.