Computer Program Turns Twitter Tweets into News Stories
A new computer program gathers Twitter trends and writes news stories.
CREDIT: Kristy Pargeter | Shutterstock.com
How are people feeling right now about Republican presidential candidate New Gingrich? Running reliable polls can be difficult and expensive, but a quick way of answering this question is to check on Tweets about him. And while a computer program automatically gathers this data, why not have it write a short news story, too? The story might look something like this:
NEWT GINGRICH GAINS ATTENTION WITH HOT-BUTTON TOPICS TAXES, CHARACTER ISSUES
Newt Gingrich received the largest increase in Tweets about him today. Twitter activity associated with the candidate has shot up since yesterday, with most users tweeting about taxes and character issues. Newt Gingrich has been consistently popular on Twitter, as he has been the top riser on the site for the last four days. Conversely, the number of tweets about Ron Paul has dropped in the past 24 hours. Another traffic loser was Rick Santorum, who has also seen tweets about him fall off a bit. (Continue reading here)
That's an entirely computer-generated, 191-word news story, written by a program from an Evanston, Ill.-based startup called Narrative Science. The company already sells its robot-written finance, sports and housing news articles to Forbes, Big Ten Network and industry publications looking for short trend reports.
Twitter reports are a logical next step. Tweets are harder for computer programs to read and analyze than the numbers in financial reports or sports statistics, but researchers around the world are working on improving Tweet-reading programs. Tweets themselves are not really news, though. Narrative Science is working on linking the tweets to current events or the status of the primaries, the company's CTO, Kris Hammond, wrote on his blog on Feb. 14. That should help the stories get past being just a recap of data.
Narrative Science programs know what is important to write about by learning concepts such as "individual effort," "team effort" or "season high," as The New York Times explained. They also know how to choose between words such as "win" versus "rout."
As for whether computer programs will replace journalists someday, Hammond once wrote that he thinks in 15 years, 90 percent of news stories will be automatically written, while tech writers have emphasized that human qualities such as "experience, judgment and creativity" are a little harder to program in.