Social media data could soon be used to identify and track people who show signs of mental illness online.
Yesterday, Minister of Science Kirsty Duncan announced more than $48 million in federal funding for 76 research teams across Canada through the Natural Sciences and Engineering Council of Canada’s (NSERC) Strategic Partnership Grants. Of that, $464,100 has been granted to Diana Inkpen from the University of Ottawa for a three-year-long project called “social web mining and sentiment analysis for mental illness detection.”
A press release issued by the university acknowledged that social media has infiltrated our daily lives. “Internet users are posting, blogging and tweeting about almost everything, including their moods, activities and social interactions,” it states. It explains how Inkpen and her team will draw upon social media data in screening for individuals at risk of mental health issues.
“We want to look at what kind of emotions people express, and then we will focus in particular on negative emotions that might show some early signs of possible mental disorders,” Inkpen told CBC News on Wednesday. “It could be depression, it could be anorexia, it could be other kinds of early mental illness signs.”
The recent launch of Facebook’s different emotional reactions in addition to the “Like” button is one way in which users’ reactions can be analyzed for trends in mood and emotion.
According to CBC, the goal of Inkpen and her team is to create a set of tools that can be used by doctors, psychologists, school counselors, and research groups to flag worrisome patterns in social media posts.
To do this, the team needs a significant amount of existing social media data to sample. They’re currently collecting data from a variety of public social media sites. Advanced Symbolics, an Ottawa-based data science technology company, has partnered with the scientists to collect such data in both English and French.
To pick up different patterns within the data and predict what these patterns mean, Inkpen’s team uses text-mining algorithms. “Expressions of very negative emotions that are very strong, or appear a lot over longer periods of time, the algorithms can pick up,” says Inkpen. “The algorithm learns from the data.”
The program can monitor how an individual’s online activities change over time. A doctor with a patient who has opted to be monitored would receive automatic alert in the event of a potential cause for concern.
Inkpen and her team plan to roll out the tool in the summer of 2018.
[ad_bb1]