At the scene of the crime: Twitter can now provide more context about this breaking news story.
Twitter has completely overhauled the algorithms and systems it uses to calculate trends as part of a recent upgrade that also saw it add more details about trending topics.
In a post on its Engineering Blog, in what Twitter describes as “the largest engineering undertaking of the trends system since 2008”, the social media giant said it has overhauled the underlying back-end platforms used to uncover trending topics.
The old trends system ran on a single Java Virtual Machine, which Twitter says could “only process a small window of Tweets at a time”. By contrast, the new platform, based on Apache Storm and MapReduce, produces far more sophisticated results in real-time.
Because of these underlying changes, Twitter will now show users more contextual information about trending topics, including a description, a photo, the volume of tweets and retweets, as well as when a topic began trending.
Twitter says the new platform allows it to automatically recompute and update this contextual information every few minutes in order to reflect “the current state of affairs”, and that it intends to “improve the quality of trends in multiple ways” in the future.
The changes to trending topics comes just days after Twitter announced a new partnership with Google that will see relevant content from Twitter appear alongside Google search results.
Andrew Sadauskas is a former journalist at SmartCompany and a former editor of TechCompany.
Comments