Twitter has taken over the social media world din the past year or two with everyone, from Lindsey Lohan to Barack Obama, “tweeting” about their day to day activities. What no one knew is that Twitter is a world of information if put into the right hands.

It had been previously used to gauge public mood, and also to predict box office sales. Johan Bollen a computational social scientist of Indiana University-Bloomington was able to do the astonishing. He used Twitter to predict the changes to the Dow Jones industrial Average with an accuracy of 86.7%.  

Bollen used a massive Google database to see how words are commonly used to express emotions. He then scanned 9.8 million tweets from 2.7 million tweeters between February and December 2008, selected the tweets that expressed emotions. He then analyzed the inputs and was amazed by the results. He analyzed twitters “mood” on important days such as the American presidential election as well as Thanksgiving. Anxiety was high the day before the election and happiness and calmness was apparent on Thanksgiving. He noticed that there was a direct correlation between calmness and the rises and falls of the Dow Jones Industrial Average. He then used a machine learning algorithm to predict whether the stock market would go up or down with the emotional data. The results were 86.7 percent accurate.

This is an incredible finding in the social media and I believe there will be an increased amount of attention focused on findings like these.

http://www.wired.com/wiredscience/2010/10/twitter-crystal-ball/