Supercomputers Can Predict Future Events

A computer with super abilities can predict the future based on data that is processed from the world's news. Research conducted in the United States shows that by incorporating millions of news related to national conditions of chaos in Libya and Egypt, a supercomputer able to predict the major events that will occur in the future.

Kalev Leetaru, from the Institute for Computing in the Humanities, Arts and Social Science, University of Illinois, USA, explained that news included in the study was taken from various sources including state-run media that the U.S. Open Source Centre and BBC Monitoring. Both monitor the news from local media across the globe.
Online news archive also analyzed such news in the New York Times since 1945. Overall, Leetaru collected more than 100 million archived news. The news was analyzed in two types of information are: mood, articles that represent good or bad news, and location, where an incident occurs and the location of the people who are in the news.
Detection mood for words such as "terrible", "horrific" or "nice". Locations where specifically collect information such as the "Cairo" and change it in the form of coordinates which is then mapped on a map. Analysis of the news elements are then used to create a network that connects 100 trillion relations.
Supercomputers are used is a SGI Altix, which is also called the Nautilus, which exist at the University of Tennessee. Computers that orchestrated the 1024 Intel Nehalem has the processing power of 8.2 teraflops (1 trillion floating point operations per second).
Based on certain data, the Nautilus can generate charts for a number of countries are experiencing what is called "Arab Spring" - a term used to describe the current media demonstration rife in Arab countries. In each case, the results are aggregated from thousands of news indicate a profound sentiment both from domestic as well as reported from the outside.
For Egypt, the tone of media coverage in the months before the fall of President Hosni Mubarak's just unheard of low-two times during the 30 years rule. Based on computer-generated graphics, Leetaru explained that while this is going something unexpected.
Leetaru research results published in the journal First Monday. Leetaru hoping his side can improve the ability of systems analysis, particularly with regard to geographical location.
"Interasi next is on stage and the surrounding cities and examine individual groups and how they interact," said Leetaru.