Since 2013, the Centers for Disease Control and Prevention’s FluSight (CDC) network has served as a kind of weather channel for influenza.

During flu season, FluSight provides weekly forecasts estimating where, when and how severely the disease will hit various parts of the country. The network, which began as a contest in 2013, is a collaboration between CDC experts and elite data analytics researchers across the nation.

In the past five years of forecasting, one team within the network stands out.

Last year, “21 different teams participated in the forecasting initiative, submitting 30 different weekly forecasts. The Delphi group at Carnegie Mellon University contributed the most accurate national-, regional-, and state-level influenza-like illness and national-level hospitalization forecasts,” the CDC says.

This marks the fourth year in a row that CMU has come out on top.

In honor of that record of ace prognostication, the CDC announced this week that CMU has been named an Influenza Forecasting Center of Excellence, a designation that comes with $3 million in research funding over the next five years.

Based out of CMU’s School of Computer Science, Delphi builds their forecasts using a combination of machine learning and input from hundreds of volunteers in the field who provide weekly reports to the team via smartphone. The team is led by Professor Roni Rosenfeld, also head of CMU’s Machine Learning Department, and Ryan Tibshirani, an associate professor of statistics and machine learning.

Rosenfeld said he was heartened to see the field of epidemic forecasting taking on a prominent place in public health policymaking: “It has become a community and more and more groups are getting involved, which is the real win,” he said.

In addition to supporting their ongoing work, the funding will enable the Delphi team to bring in collaborators from outside the university. A team from Harvard’s T.H. Chan School of Public Health will work with CMU to apply their research to the study of pandemics, while researchers from the University of Pittsburgh’s Graduate School of Public Health will collaborate on data gathering and analytics.