Who needs an awards show to tell us what movies and actors fans prefer when we have Twitter? In yet another partnership with the USC Annenberg Innovation Lab, IBM is turning its skills in social-media sentiment analysis to Hollywood awards so the world can see which movies and stars are generating the most buzz on Twitter. IBM has done similar analyses of both the World Series and the Super Bowl, and although they’re no doubt part of a marketing effort to demonstrate its big data prowess, the projects are pretty fun and rather insightful.
It’s new Senti-meter, which appears as part of the Los Angeles Times‘ interactive and ongoing awards section called “The Envelope,” ranks movies, actors and actresses based on the number of tweets about them and the sentiment contained in those tweets. For example, as of its last update on Jan. 28, the Senti-meter showed Hugo dominating in number of tweets, but Midnight in Paris generating the most-positive reaction. The Girl With the Dragon Tattoo, however, appears to have the best balance between number of tweets and positive tweets.
It’s easy to take fan sentiment with a grain of salt if you’re a movie critic, but if you’re a studio, being able to quantify the types of movies and stars fans prefer could mean big bucks. When I spoke with IBM SVP of IBM’s Software and Systems Steve Mills in October, he told me IBM expects to do $16 billion in analytics revenue by 2015, and he expects social-media analysis to be a big driver of that growth.
Anyone into betting on the various awards shows might find this particular project useful, too. It’s hardly predictive analytics, but the masses might know something: although the New England Patriots were favored in last week’s Super Bowl, sentiment swayed in the last few days to New York Giants quarterback — and eventual winner — Eli Manning from Patriots quarterback Tom Brady.