App could identify depression just from social media updates

Giving out personal information on Twitter may soon be positively encouraged, as scientists have looked into developing a tool that could identify whether a person is depressed just by looking at their tweets, reports.

Over recent years, health professionals have taken statistics on depression with a pinch of salt, arguing that many more cases may go undiscovered - and therefore untreated - by people who do not seek help. This may all be set to change, though, thanks to pioneering tech work from researchers at the Centre for Statistics and the Social Sciences at America's University of Washington.

Looking at something as simple as tweets, the researchers argued, could give a clear indication as to whether or not an individual is suffering with depression. It doesn't involve developing an app that picks out certain key words or phrases, such as 'sad' or 'depressed', however, but instead one that takes into account the volume of messages sent and time of day in which the individual was active.

Learning a person's mood from these updates, researcher Eric Horovitz told, should then be a relatively straightforward process, even potentially highlighting issues before they take hold. All that was needed was the capacity to "build measures that might be able to detect if someone is severely depressed, just in publicly posted media.

"You might imagine," he added, "tools that could make people aware of a swing in mood, even before they can feel it themselves."