• by Erik Sass , Staff Writer @eriksass1, March 3, 2016

In one short decade, social media has accumulated the biggest pool of information on human sentiment and emotion ever known, and we’re only beginning to explore the potential of this huge, constantly growing database in areas ranging from marketing to social science and mental health. On the latter note, the Canadian government is funding researchers who hope to use social media to identify mental illness.

This week researchers at the University of Ottawa and elsewhere received a half-million dollar grant to study “social web mining and sentiment analysis for mental illness detection.” The team, which also includes scientists from the University of Alberta and the Université de Montpellier in France, will use algorithms to sift social media data to detect individuals at risk of major mental illness or already suffering with it.

As part of the complex, three-year long study, the team will partner with a company, Advanced Symbolics, to collect and sample social media data using medical informatics and natural language processing, ultimately developing algorithms correlating linguistic clues with various types of mental illness.

According to the University of Ottawa, the same techniques can eventually be applied to detecting youth at risk of delinquency and victims of high school bullying.

This is just the latest in a series of research programs aiming to mine social media for clinical purposes. For example, last year GlaxoSmithKline revealed that it has collected a trove of social media data consisting of millions of posts about hundreds of drugs, including drug interactions.

GSK worked with Epidemico, a technology company focused on healthcare insights owned by Booz Allen Hamilton, to surface and analyze 21 million posts mentioning its product on Facebook and Twitter, including 15 million posts on Facebook alone.