Researchers training app on when to notify users

Mar 24, 2017, 7:58 AM EDT
(Source: Matthew Hurst/flickr)
(Source: Matthew Hurst/flickr)

Janne Lindqvist and his team at Rutgers University in New Jersey don’t want smartphone users to be passive ducks to the barrage of app notifications that keep flooding through the day with no sensitivity to the person’s state of mind or schedule. Their solution is a machine learning app that gathers data related to a person’s location, current activity and personality to time the notifications for “least interruption.”

Matthias Böhmer at the Technical University of Cologne, Germany, admits that a number of app notification filtration systems already exist but their “broad range” doesn’t fit all situations. The new app would address this gap, for example, a number of non-urgent notifications can be held for certain hours if a user is busy, say in gym, a luxury that no existing system or device offers.

Experts believe the new app would actually be a tool in the hands of app creators, who could target the users in those hours when chances of getting a response are high, writes New Scientist.