Last week at the Time Warner Center it was announced that CNN and Twitter had Partnered with Dataminr for News. I’ve linked to a couple of the articles that speak about the product. For me this was one of the most exciting products that I have worked on so far, from the process start to finish. Taking unstructured data and making something incredibly helpful on multiple levels, designing around a lot of different formats and all the UI challenges that came along with real time info display. On top of all that, just seeing news being made as it is happening is incredibly powerful. Along with the excerpts from the articles I’ve posted are a couple sketches and images from the video presentation.
By the time a topic is “trending” on Twitter, it’s probably old news already. Today in New York City, data-crunching company Dataminr announced a new tool for journalists. Its goal is to seek out news stories before they’re heavily reported.
Dataminr For News does this by scanning Twitter, albeit in a slightly different way from existing filter tools. The news-gathering tool was developed as a partnership with Twitter and CNN, whose reporters and editors have been using the software for six months and have helped shape it during that time.
The company says it now produces about two stories a day based off tips from its Dataminr alerts. “It’s like bionic vision for our reporters,” said Kenneth Estenson, general manager of CNN digital. “It helps us to see things faster than our competition and to act with confidence.”
Dataminr CEO Ted Bailey said the goal is to “alert journalists to information that’s emerging on Twitter in real time.” Basically, the technology looks at tweets and finds patterns that can reveal breaking news when it’s still in its “infancy.” Those alerts can be delivered in a variety of ways, including via desktop applications, email, mobile alerts, and pop-up alerts.
The program, which was piloted with CNN, isn’t just for alerts, Bailey added. He said it also offers analytics data around who broke the news and how it spread. He also said that it doesn’t just have to be for breaking news, but for finding interesting features, as well.
Dataminr scans Twitter for patterns based on a variety of factors including language and location, which are then put through algorithms to select important events and present them to journalists as alerts. The company previously specialized in providing financial analysis of sentiment on Twitter. It rose to prominence by detecting the impending announcement of Osama bin Laden’s death before the major news outlets.
CNN president Jeff Zucker lauded the partnership as “a really improtant tool and resource as we continue to gather news,” in comments at Wednesday’s event. “There have been many examples already of how we’ve used Dataminr.”
TweetDeck is already an insanely helpful tool for journalists, but it’s only useful if you know what you’re looking for. Enter Dataminr for News, a soon-to-be released product that is the result of a partnership between CNN and Twitter. The idea is to help journalists get the most out of Twitter by alerting them of emerging information in real time. Dataminr also offers analytics around who broke the news, and how it spread on Twitter.
In order to distinguish between accurate information and what just might be a hoax, Dataminr takes into account 25 different things to determine how legitimate the tweet is, including user influence, geolocation and momentum.