Next time you’re in New York City and wondering where the party’s at, or if you’re going to get stuck on 8th Avenue in traffic from the Rangers’ game, you might do well to consult an unlikely ally: Twitter. Scientists have analyzed 6 million geolocated, time-stamped tweets from New York City and the surrounding area and discovered a “heartbeat” that says a lot about how New Yorkers live. As seen above, the frequency and location of tweets swells and recedes rhythmically, with red representing higher than average use and blue representing lower than average. Unsurprisingly, the study, published today on the arXiv preprint server, found that the most prominent predictor of Twitter usage is our daily sleep cycle. But by analyzing the locations of the tweets, the team was also able to watch New Yorkers commute from the suburbs into Manhattan, gather at stadiums for sports events, head downtown during late-night hours, and more. The team also found that Twitter usage tended to increase in transportation hubs where users are presumably bored with nothing else to do. The scientists posit that analyzing large-scale social population dynamics could help us design safer and more efficient cities—a goal that anyone who has ever tried to squeeze into an F train in midtown during rush hour can certainly appreciate.
(Video credit: U. França, H. Sayama, C. McSwiggen, R. Daneshvar and Y. Bar-Yam, Visualizing the “Heartbeat” of a City with Tweets. New England Complex Systems Institute [21 May 2014])