The chatter in Twitter can accurately predict the box-office revenues of upcoming movies weeks before they are released. In fact, Tweets can predict the performance of films better than market-based predictions, such as Hollywood Stock Exchange, which have been the best predictors to date.
The Social Computing Lab at HP Labs in Palo Alto, CA found that using only the rate at which movies are mentioned could successfully predict future revenues. But when the sentiment of the tweet was factored in (how favorable it was toward the new movie), the prediction was even more exact. To quantify the sentiments in 3 million tweets the team used the anonymous human workers found by the Amazon Mechanical Turk to rate a sample of tweets, and then trained an algorythmic classifier to derive a rating for the rest.
The graph above compares the predicted vs actual box office scores of tweet-rates (blue line) and Hollywood Stock Exchange (green line).
Benardo Huberman, the chief investigator on this work, says that they predicted the outcomes of new movies released in November and December 2009 and January 2010, including Avatar, Invictus, The Blind Side and Twilight.
Of course, predicting movie revenues is only a tidy test case. If you can use Twitter to predict the future of movie tickets, then why not elections, or sales of other products? As the authors write:
This method can be extended to a large panoply of topics, ranging from the future rating of products to agenda setting and election outcomes. At a deeper level, this work shows how social media expresses a collective wisdom which, when properly tapped, can yield an extremely powerful and accurate indicator of future outcomes.
The PDF version of the paper Predicting the Future With Social Media by Sitaram Asur and Bernardo Huberman is short and clear.