BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

Here's What 6 Billion Tweets Looks Like Mapped

This article is more than 9 years old.

What do you do with three terabytes of geotagged tweets? You map them if you’re a cartographer like Eric Fischer.

A data artist and developer at Mapbox, Fischer has been collecting geotagged tweets for the last three and a half years using Twitter’s public API. With more than 6.3 billion tweets (6,341,973,478 to be exact) in his database, Fischer made an interactive map that’s detailed to street level.

Ultimately, only nine percent of the six billion tweets were represented as dots on the map. This is due to filtering that removed duplicate coordinates, mapping unique latitude and longitudes only once on the map.

Interestingly, with or without filtering, a large visible stripe that’s apparently devoid of tweets appears over the prime meridian in London. Fischer suggests in a blog post that Twitter is likely responsible for filtering the tweets in that area, for reasons not known.

Here’s a screenshot of John F. Kennedy International Airport near New York City. Zooming in to see the individual dots, you can see that some gates tweet more than others.

Sporting events also generate a lot of tweets. But because the map doesn’t allow for a user to search for a particular address, I chose to look up Target Field in Minneapolis – the home of the Minnesota Twins and a stadium I’m familiar with, having gone to school in the Twin Cities.

From a zoomed-out view above Minneapolis, Target Field is lit up with dots, but zooming in on the ballpark, a majority of the dots are shown on the playing field, with a high concentration over second base.

Surely Minnesota’s second baseman has better things to do than to tweet during games, but there are a few other logical explanations to the data. For one, Target Field isn’t just a baseball stadium; it’s also a concert venue that’s seen the likes of Paul McCartney and Kenny Chesney in recent years.

The most logical explanation, however, could be the coordinate data itself. As Fischer acknowledged in his post, tweets from iPhones seemingly blur latitude and longitude intentionally to avoid giving away a user’s exact location. So what may seem as tweets sent from second base could just be tweets that sent Twitter the general coordinates of Target Field.

The data that Fischer used to create his map is accessible to anyone with the know-how of using the Twitter API. Fischer also provides links to the code he used to filter the data as well as instructions on how to map the data in case anyone wants to recreate a similar map.

Follow Frank Bi on Twitter at @FrankieBi