Pandemics hold an alarming possibility with the last major one, the 1918 Flu Pandemic claiming more than 50 million lives. Of course in this modern day and age, with modern means of travel enabling viruses to spread even quicker, the possibility of a pandemic is even more alarming and with recent scares such as SARS, Swine Flu and Bird Flu, pandemics almost seem like realities. Of course while it may be modern conveniences that could aid the spread of diseases it is the very same conveniences that could help stop them as well.
While Twitter may be known as the most popular micro blogging website it can in fact have other purposes, with researchers suggesting that it be used particularly in the case of a viral outbreak. Researchers from the University of Rochester, New York have proposed that the micro blogging website could be used in case of a endemic or pandemic to ‘track’ the spread of a disease and even possibly tell when exactly it may affect you.
Speaking to the magazine New Scientist at the Conference on Artificial Intelligence in Toronto, Canada the researchers explained how they used Twitter to essentially plot a map that could show viral hot spots and indeed show the progression or spread of a disease. In the present study, the researchers tracked the spread of the common flu across New York, using tweets from those who were sick or who had the symptoms to make a ‘heat map’ of New York, showing the where the flu was concentrated and how it was spreading.
The university of Rochester team used 4.4 million GPS tagged tweets from around 600,000 users in New York City for a period of one month to chart their map, using an algorithm to sift between the various tweets to identify those who were genuinely sick. Speaking about the study, Adam Sadilek of the University of Rochester said, “Our models enable you to see the spread of infectious diseases, such as flu, throughout a real-life population observed through online social media. We apply machine learning and natural language understanding techniques to determine the health state of Twitter users at any given time. Since a large fraction of tweets is geo-tagged, we can plot them on a map, and observe how sick and healthy people interact. Our model then predicts if and when an individual will fall ill with high accuracy, thereby improving our understanding of the emergence of global epidemics from people's day-to-day interactions.”
According to the researchers, the algorithm they had developed was accurate 90 per cent of the time and around ‘8 days in advance,’ with Mr Sadilek adding, “We show emergent aggregate patterns in real-time, with second-by-second resolution. By contrast, previous state-of-the-art methods (including Google Flu Trends and government data) entail time lags from days to years.”