After the fall of Google Flu Trends, researchers suggest Wikipedia could take it’s place
Nature recently reported on research from the Massachusites Institute of Technolohgy (MIT) that Google’s tool for predicting the number of flu cases, Google Flu Trends, was consistently over estimating the number of cases. This was put down to changes to the google search algorithym, and the introduction of suggested search terms. This was a blow for what many said was a pioneering use of big data to predict and inform planning for seasonal events.
But now David Mclver and John Brownstein from the Boston Children’s Hospital have suggested that public health officials could use Wikipedia to predict seasonal flu trends. By applying the same principles as those behind Google flu Trends, Mclver and Brownstein believe that analysis of American internet traffic to influenza-related pages on the worlds largest encyclopaedia could prove to be a more accurate predictor of seasonal flu patterns.
In fact, their paper, published in PLOS Computational Biology, the two researchers showed that analysis of Wikipedia traffic was 17% more accurate than Google Flu Trends between 2007 and 2013.
“Each influenza season provides new challenges and uncertainties to both the public as well as the public health community… We’re hoping that with this new method of influenza monitoring, we can harness publicly available data to help people get accurate, near-real time information about the level of disease burden in the population.”
Find us on Google