Detecting breaking and hot news in real time. NewsFlow: the evolution of news
Many news articles are published on a daily basis. And yet, in hindsight, some of them are not particularly relevant and quickly fade away, whereas others are of utmost importance. Moreover, some pieces build on previously published ones, forming relationships that might be entangled and complex. Information is power, so how do we separate the wheat from the shaft in a way that is painless and quick?
At NaturalTech, we’ve pondered how our cutting-edge technology might help us in this quest. We wanted to show you which articles from different media outlets are related, in the sense that they cover the same event, and how relevant the event is because it has a lot of repercussion in the media.
This is how we came up with NewsFlow: NaturalTech’s powerful tool that allows you to visualize the relationship among news and how this relationship changes over time. How does it do it? We have created a video so you can understand the magic behind NewsFlow:
In the video, you can see, so to speak, the life cycle of a news article, T0. When it is first published, and NewsFlow doesn’t detect existent relationship with any previously published article, T0 is marked as trivial.
However, trivial news has the potential to become breaking. This happens when another news article, T1, is published soon after T0 and NewsFlow detects that both articles are related. In this scenario, we say that T0 is the daughter of T1 and, T1 the mother of T0. Hence, T0 becomes breaking news, and T1, regular news. See in the example below how the most recent article comes in (T1) and a relationship with a previous article is established. Once this happens, T1 is labelled as Regular News and T0 as Breaking News:
But this is not all. If two new articles are registered (T2 and T3), and they both have a relationship to T0 and T1, T2 becomes Hot News and T3 Super hot news.
NewsFlow shows you the news up to the last 30 days, both in Portuguese and Spanish, with their source, relevance label and a NewsTree showing previous articles that are related. It also lets you filter the categories that are most relevant: