In addition to connecting people and (maybe) serving as an advertising platform, social media can be used to predict a variety of phenomena, according to scientists who are mining the huge volume of conversations and opinions posted online and linking this data to real-world outcomes.
One recent example comes from physicists at Japan’s Tottori University, who say they’ve come up with a model which correlates a number of factors, including social media buzz, to the box office success of new movie releases. The model also takes into account the amount of money spent on advertising a movie before its release, and the length of time devoted to the campaign.
The model and some successful predictions are summarized in an article, “The ‘hit’ phenomenon: a mathematical model of human dynamics interactions as a stochastic process,” published in the New Journal of Physics. According to the researchers, the model was able to predict the popularity of movies including “The Da Vinci Code,” “Transformers,” “Spider Man 3,” and “Avatar.” So far the model has only been applied to the Japanese market, but the authors claim it can be applied to any market.
Discussing the word-of-mouth element of the model, including indirect communication and rumors, the authors note: “We found the daily number of blog posts to be very similar to the revenue of the corresponding movie. The daily number of blog posts can be used as quasi-revenue. The results calculated with the model can predict the revenue of the corresponding movie very well.”
As noted, other scholars are also coming up with predictive models based on social media. In March I wrote about research by SAS and UN Global Pulse, a UN think tank, drawing on social media conversations in the U.S. and Ireland from June 2009-June 2011 to predict economic trends including unemployment. The project, titled “Unemployment Through the Lens of Social Media,” set out to compare qualitative information gathered from social media with unemployment figures. The researchers examined online job-related conversations from blogs, forums and news in the U.S. and Ireland and assigned a quantitative “mood score” based on the tone of the conversations.
The quantified mood scores were then correlated to the unemployment rate, revealing leading indicators that are able to forecast rises and falls in the unemployment rate. For example, the researchers found that the volume of conversations in Ireland showing a “confused” mood correlated with an uptick in unemployment three months later. Likewise, conversations about public transportation spiked about a month before unemployment.