Software might not be able to think, but its writing is getting stronger by the day. Narrative Science, which creates automated stories based on data sets like sports statistics or earnings, has turned its hand to "unstructured" data like lists of Twitter messages. In its latest project, it set its tools to capturing and tagging information about presidential candidates, then automatically generated a story based on the result. Below is a sample from the piece, entitled "Newt Gingrich gains attention with hot-button topics taxes, 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."
The article also incorporated pertinent quotes, like one "tweeter" who spoke favorably about Gingrich's budget proposal. There's some awkward phrasing in the piece, but it's otherwise passable prose, especially considering that as CEO Stuart Frankel says, "no human touched it at all." The process is still in beta; by the time it's done, it'll likely start showing up in things like private briefings, which require good information more than stellar turns of phrase. From there, though, it's not a huge step to having these automated processes write up low-level analysis on public-facing sites, much as Forbes has already done for earnings previews.
We'll email you a reset link.
If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.
Choose an available username to complete sign up.
In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.