University of Maryland Machine Learning researchers developed a predictive algorithm to enhance Facebook’s notification system for blood donations. This algorithm prompted 5% more people to take blood tests in the last year, which could mean 140 million more potential donations worldwide if applied to the 2.8 billion users on the social media platform.
Details of this research were presented at the Association for Computing Machinery’s Conference on Economics and Computation.
‘This is a successful step in better aligning the global supply and demand of blood,’ said co-author John Dickerson.
Over than 4.5 million Americans require a blood transfusion annually, with an even greater demand in poorer countries. To assist in increasing the supply, Facebook created a tool which notifies potential donors of opportunities to donate. Since the tool’s kickoff in 2017, over a 100 million people in 27 countries signed up to receive blood donation notifications.
Facebook collaborated with University of Maryland Center for Machine Learning researchers to enhance its notification process.
Instead of just sending random notifications of close-by giving opportunities, the team designed special algorithms to detect when and where users were most likely to donate. This was dependent on factors such as age, proximity to donation locations or how often they used Facebook.
The team compared the predictive model to one that chooses nearby giving opportunities randomly. 1.5 million Facebook users received a notification during the pilot experiment. The team’s predictive model performed better than the random model, with about 75,000 more donations.
According to Duncan McElfresh, the lead author, although the potential impact is impressive, there is still massive room for improvement.
By Marvellous Iwendi.
Source: University of Maryland