WEST LAFAYETTE, Ind. (AP) – A team including researchers from Purdue University has made a computer model that tracks the flu across the U.S. using Google searches and tweets.

The (Lafayette) Journal & Courier (http://on.jconline.com/1dMZ4dH ) reports the model was developed by researchers from Purdue and Indianapolis-based startup Perscio. The model uses Google and Twitter data to see how many people are talking about the flu and where.

The predictive tool also uses transportation, weather and population data in its measurements. It can make predictions on flu trends as many as five days into the future.

The team won $20,000 recently as part of IBM's Big Data for Social Good project.