Sickness Forecast With 90% Accuracy by Twitter

Twitter can predict with almost 90% accuracy about the likelihood of you falling ill, say researchers from the University Of Rochester in New York. Twitter was able to track the outbreak of flu around a certain area by studying 4.4 million tweets from 630,000 users that had the GPS location data.

Using an algorithm which could perceive the difference between actual sickness and usage of the common word ‘sick’, researchers were able to plot and predict when people from a certain area were at risk of picking up a flu bug and the good thing is that twitter can inform you about the risk a week in advance.

The algorithm is based on a model that analyses the spread of infectious diseases among a population, based on twitter messages received and concentration of so-called ‘unwell’ messages in a certain area.



Source-Medindia