Find a Question:
Scientists use Wikipedia for predicting flu
Scientists at the US Los Alamos National Laboratory have devised a method of predicting by means of the Wikipedia when influenza prevails in a certain area. The idea is reminiscent of Google flu prediction tool Flu Trends.
The researchers took part in a competition for a Center for Disease Control and Prevention, a regional institution in the United States that are in a state concerned with the detection, treatment and prevention of disease. The intention was that the participants with data available on the Internet, trying to predict when the flu would break. The Institute is namely jump to a new measuring method. Is the current which is to be limited, since it is based on counts of volunteers.
The scientists at Los Alamos National Laboratory used their method real-time statistics of Wikipedia pages about flu and flu symptoms. They chose those dates because they are freely available. Through machine learning and logs from previous years, the scientists wrote a program that learned to recognize patterns in the outbreak of the flu.
The experiment showed significant that the program can predict when the flu breaks out, writes MIT Technology Review, based on the research. The numbers derived from the program were were able to model the flu data of the Center for Disease Control and Prevention, which is the center two weeks later published, on the basis of the measuring method with the volunteers.
Perfect is the only method still far, the researchers emphasize. The program can include not ‘see’ when an epidemic is over, because it is difficult to recognize. This is necessary: ’peaks’ of future virus outbreaks based the script on data collected in previous years. In addition, the program can not yet detect two or more simultaneous outbreaks.
Incidentally, it is not the first time that real-time data will be used for the scientific prediction of virus outbreaks. Among other things, Google has been doing that for years with his prediction tool flu Flu Trends. The algorithm of the search giant looks at how often searches for flu symptoms. These data link to a Google site, and see where the flu would prevail.
The reliability of Flu Trends late to still be desired: the forecast in 2012 and 2013 week up to 95 percent. That was mainly because Google showed higher numbers of flu then see there were actually known. The search giant says for future ‘flu season’ to use a new algorithm, but has yet no technical details were disclosed.Viewing:-198
Answer this Question
You must be Logged In to post an Answer.
Not a member yet? Sign Up Now »
Star Points Scale
Earn points for Asking and Answering Questions!