Twitter is one of the most famous social websites of the world and has been used for a number of purposes such as planning social lives, interacting with the celebs and communicating with friends. Now twitter is getting into a serious business to help track diseases. This technology of tracking diseases has already been used by researchers to track flu as it spread around New York with the aid of heatmap of the users who whine of suffering from illness.
Adam Sadilek at the University of Rochester along with his team examined about 4.4 million tweets from 600000 New York users over the period of one month. The artificial intelligence was trained in such a manner by the team that it was able to ignore all the tweets from healthy people and find the people who were actually suffering from illness. According to Adam Sadilek friendship is the key behind this system.
The system works in this way: – Suppose there are three friends of yours who suffer from flu and you also met about eight people who have symptoms of such flu like headaches or runny noses then what is the possibility that you might also get ill? The models would help the user to see the spread of flu throughout the population in real life events based on the media reports. All the tweets are used to plot a graph that would help a lot in predicting that in how much danger the particular user is and when he has the highest risk of getting ill. Machine learning as well as natural language understanding technique has been involved to determine healthy condition of the twitter users.
Sadilek also said that because most tweets are geo tagged it becomes very easy to plot them on map and also observe the interaction between healthy and ill people. Based on this, the algorithm would predict if as well as when the particular individual shall suffer from illness and thus help in increasing the understanding of global epidemics and how they spread form person to person through their interaction.