Better track geography and know where stories are being published and talked about
Going forward, this study provides an interesting foundation for thinking about how our media are interrelated, and how various facts, anecdotes, and bits of misinformation make their way to the public.
"Can we start exploring the data not from identifying these topics of keywords upfront, but asking an algorithm to surface some of those for us?" asks Graeff. "What are some unusual things or clusters of news stories that will allow us to get a sense of news stories that otherwise wouldn't be seen?"
In the future, Graeff says he'd like to be able to better track geography and know where stories are being published and talked about. The team is also interested in using natural language processing to track the spread of quotations from source to source. In addition, "automated coding and sentiment analysis" could be used to better understand how perspectives in the newsroom are molding stories -- tools like OpenGender Tracker.
Ultimately, the goal is to create a suite of tools that activists, journalists, and academics can learn from. Says Graeff: "A lot of what we show here is that there are better methods for studying the media as a so-called media ecosystem that allow us to really understand how a story goes from barely a blip to a major national/international news event, and how controversies circle around that."