Social media data used to predict depression in users

4 August 2014

A new study has found people who suffer from depression have a higher activity on social media between 11pm and 3am compared to other social media users.

The joint study from The Australia National University (ANU) and China’s Harbin Institute of Technology (HIT) analysed data from Chinese social media site Weibo to predict potential depression among users.

“We found that in comparison to the control group, individuals with predicted Major Depression Disorder were more active on social media between 11pm and 3am,” said co-researcher Jin Han, a PhD student with the ANU National Institute for Mental Health Research.

“Secondly, these people have a specific linguistic pattern. They are more likely to use negative words, such as death, depression, life, pain, and suicide.”


Ms Han said she hoped the study could be used to help organisations which work with people suffering depression.

“The information could be provided to organisations such as Lifeline who support people with depression,” she said.

“It could also be used to identify people who may be showing signs of depression but are less likely to reach out for help.”

Ms Han presented the findings of the study during the two-day Studying the Chinese Internet workshop jointly organised by the ANU College of Arts and Social Sciences and the ANU College of Asia and Pacific.

The other researchers are Pengyu Li, Dr Guang Yu, and Xianyun Tian from HIT.

Workshop organiser Associate Professor Robert Ackland said the event had been successful in bringing together people who work at the intersection of computer science and China studies.

“One of the things the ANU is known for is its focus on Asian Studies, and I think we are very well positioned to be a leader in this field of study,” he said.