When associate professor of informatics Johan Bollen began research on predicting trends in economic markets using Twitter in 2010, he was driven by curiosity and a passion for learning.
Since publishing his work with co-author and graduate student Huina Mao, their research has received an official United States patent and led to the creation of a new technology startup company, Guidewave Consulting.
“As soon as we published the paper and had it submitted for a patent, we got a lot of interest from a variety of potential licensees, hedge funds, banks, you name it,” Bollen said. “But we didn’t have a vehicle to leverage that opportunity.”
Instead of selling the technology to the University, Bollen wanted a direct say in how his mood-tracking system would be marketed to investors. With the help of IU Research and Technology Corporation — a current shareholder and investor in Guidewave — a percentage of Guidewave revenue is collected by the University, allowing both Bollen and his employer to benefit from IU-sponsored research.
“It’s everyone making sure that if this thing really takes off, that IU and the state of Indiana can profit from it as well,” Bollen said.
When Bollen and Mao published “Predicting Economic Trends via Network Communication Mood Tracking” in 2010, media affiliates such as CNBC quickly picked up on the data-analysis technology, deeming it the “Twitter Predictor.”
The series of algorithms designed by Bollen and Mao categorize Twitter updates from millions of users into different “moods” such as tension, depression, fatigue and vigor and then uses changes in these moods to compare them with trends in economic markets such as the Dow Jones Industrial Average.
“A lot of people think these social media environments are just kids posting crazy stuff,” Bollen said. “But it could have tremendous value in predicting social and economic trends. There’s a little bit of disdain for these social media at times, and to some degree it’s justified. But the scale is enormous.”
While the correlation between mood on Twitter and short-term economic trends is still widely debated, Bollen said Guidewave Consulting has been in contact with a variety of interested hedge fund investors and marketing researchers since the company’s founding in June 2011. Bollen even noted that undisclosed customers have purchased their research data and used it in market analysis.
“The nice thing about this is you don’t need a huge team to do this type of analysis,” Bollen said. “The computers do most of the work. It’s all about having smart algorithms and applying them intelligently to a rather difficult problem.”
Guidewave currently operates with three employees: Bollen, Mao and graduate student Damien Junk. All three, even Bollen as founder and CEO, are by no means full-time employees. Each continues to research full-time at IU.
“I think I’ve aged badly over the last two years,” Bollen said. “But IU is a pretty receptive environment to these kinds of things. It’s been very rewarding.”
Though Bollen said Guidewave is still in a developmental stage, one or two significant customers could jump-start revenue.
“This is the kind of business where a few contracts can make a big difference,” Bollen said. “We’re not looking to take this into the retail space. We’re gearing this toward pretty big investors, like really large hedge funds. You can go from revenue-neutral to very high revenues in a very short amount of time.”
With a recently patented invention and a small team of dedicated employees, Bollen said he is optimistic about the future of Guidewave but emphasizes he is a professor before an entrepreneur.
“We’ve got a startup, and it’s real, but it’s not like we’re hiring 150 people,” Bollen said.
As much as he enjoys developing a new company, he said his true passion still lies in the excitement of academic research and discovery.
“It’s fun to take a piece of research and turn it into something that works in real life,” Bollen said. “Curiosity and fun are really powerful motivators for me. As long as it’s fun, I think it’s valuable.”