Supercomputers could predict major upheavals with some degree of confidence, suggests Georgetown University researcher Kalev Leetaru. He analyzed more than 100 million articles from media outlets around the world, spanning 30 years, with a shared-memory supercomputer called Nautilus.
The findings point to a degree of predictive ability greater than chance would account for. For example, the revolutions in Tunisia, Libya, and Egypt, including the removal of Egyptian President Hosni Mubarak, were among the events predicted, as well as a period of instability for Saudi Arabia.
However, the real test will be anticipating events that have not yet happened. Supercomputers should be given a chance because humans are even worse at predicting global conflicts, according to Leetaru, formerly affiliated with the Institute for Computing in the Humanities, Arts, and Social Science at the University of Illinois.
He compares the predictions to weather forecasting. "A 70 percent chance of rain tomorrow means that it might not rain, but it's probably worth bringing an umbrella, because the conditions for rain are there," Leetaru says.
From HPC Wire
View Full Article
Abstracts Copyright © 2013 Information Inc., Bethesda, Maryland, USA
No entries found