AI Journalism: How Reuter’s Tracer Works in News Room

At LightRay Media, we were fortunate to be selected to join the London School of Economics Discovery AI Journalism supported by Google News Initiative.

One of the great AI tools we were recently introduced to applied in big news rooms like Reuters, is its News Tracer. It is an idea that small newsrooms can incorporate into their media development products, services and strategies.

Reuters News Tracer

Over a two year period, Reuters and technology professionals have been developing a solution: Reuters News Tracer™. This tool has enabled journalists to spot and validate real news in real time on Twitter® – Reg Chua Executive Editor for Editorial Operations, Data & Innovation, Reuters

Harnessing the power of the crowd, Reuters News Tracer receives alerts that enable it to tap into worldwide eyewitnesses, to see what’s happening around the world. Crucially, this is providing more time to do value-added reporting work.

With fake news challenging the veracity of news and integrity of information, Reuters has developed a tool that is combatting the problem and providing its journalists anywhere from an 8- to 60-minute head start. Increasingly, events surface first on social media as people post what they’re seeing, hearing and experiencing in the moment. With the proliferation of smartphones and social media, there are many more eyewitness accounts of a lot more events.

However, it is the veracity of news and the integrity of information and sources that have been making headlines of their own lately. So how does Reuters identify and verify the real stories in the vortex of social media chatter?

How does it work?

Reuters News Tracer uses the power of cognitive computing and machine learning to extract insight from the immense stream of social media. The tool combines algorithms which merge artificial intelligence with the human intelligence of Reuters journalists.

First, it runs machine-learning algorithms on a percentage of Twitter’s 700 million daily tweets to find breaking news.  These algorithms look for clusters of tweets that are talking about the same event and the tool then generates a newsworthiness rating, questioning whether the event is worth reporting.

It then “thinks” like a journalist through reverse-engineering how a journalist would verify whether a piece of information was true. It delves into the identity of the twitter profile, checking whether it’s a verified account, whom the profile they follow, who follows them, whether the tweet contains links and images, and the structure of the tweets themselves – along with many other factors.

Reuter’s journalists then independently verify the information through their own channels and reporting, before publishing. The human expertise of Reuters journalists – and judgment of their peers across the world – forms one of over 700 signals the algorithms read from a tweet to determine its veracity.

The capabilities of Reuters News Tracer have been covered by Harvard’s Nieman Journalism Lab and featured in the Columbia Journalism Review. Its state-of-the-art technology resulted in six patent filings and 20 published papers.

The results

Reuters News Tracer has detected numerous breaking news stories since its distribution.

It detected accounts of a shooting in San Bernardino, California, before any major global news organization. When an earthquake in Ecuador caused the deaths of 77 people in April 2016, Reuters News Tracer gave its journalists 18 minutes to gather more information before another news outlet broke the story. It also helped the news team get an 8-minute head start in reporting on the Brussels bombings, and a 15-minute head start on sending out a news alert on the Chelsea bombing in New York in October 2016.

The capabilities of Reuters News Tracer have been covered by Harvard’s Nieman Journalism Lab and featured in the Columbia Journalism Review. Its state-of-the-art technology resulted in six patent filings and 20 published papers.

Read More Insights 

A version of this article was originally published with the Thomson Reuters annual report.

https://www.niemanlab.org/2016/11/reuters-built-its-own-algorithmic-prediction-tool-to-help-it-spot-and-verify-breaking-news-on-twitter/