Analysts have been studying social media for patterns that emerge across it. Most notably, news of illnesses and diseases are often cited on social media sites and thus, a region-wise or time-wise pattern can be established by reading, say, the tweets on Twitter. Graham Dodge, founder of Sickweather, has been able to discern the fastest disease spreading regions in US by analyzing social media streams.
Dodge has noticed that diseases tend to spread most fast in the region between Hartford, Conn and Washington D.C. Due to this, he terms this area ‘contagion alley.’ A very significant factors that adds to the spread of diseases is huge events, such as the Super Bowl. This February, for instance, when the Super Bowl was due, different kinds of illnesses in Indianapolis were discussed nearly twice as many times as is normally done on Twitter.
Earlier, social analysts had hoped that they may be able to track diseases in different regions with the help of social media. Now, the hopes are even higher – analysts are hoping that they may be able to predict disease outbreaks using different disease patterns on social media in the coming days.
Sickweather is an ambitious venture by Dodge. It essentially attempts to track diseases with the help of social media. So how does it work? It scans multiple social networks to discern mentions of illnesses. It then analyzes these mentions to see if they actually refer to an illness or are merely metaphoric. Once they are found to be relevant, they are then plotted to the region that tweet originated from.
In this way, Sickweather is able to create a map which can show disease outbreaks according to location and illness. Social media has come about to be of great value in this way. According to Dodge, “It inherently provides more context to the individual’s situation for natural language processing to better qualify what the person means, e.g. a tweet of ‘I have the flu’ versus a search of the word ‘flu.”
Naturally, analysts have tried plotting disease patterns in the past through other, different indicators. Now, these analysts hold that utilizing social media is the fastest way. And these analysts are very hopeful that in the coming days, disease predictions are going to further improve. Dodge says, “Think of it like the early days of weather forecasting,. We’re basically transitioning from the Farmers’ Almanac to Doppler radar.”