Is Boston’s Beacon Hill or Cape Town’s Bo-Kaap livelier? Does Somerville or Sydney look safer? A new MIT Media Lab project, powered by Google Street View and hundreds of thousands of clicks is trying to find out, testing the limits of what crowdsourcing pictures can tell us.
The project, dubbed Place Pulse, shows the visitor two random pictures culled from locations in a number of cities on Google Street View. The user is presented with a simple question (Such as “Which place looks livelier?” or “Which place looks safer?”) and told to click one of the pictures or to pronounce a tie.
“Place Pulse aims to quantitatively recognize which areas of a city are perceived as wealthy, modern, safe, lively, active, unique, central, adaptable or family friendly,” the project’s website states. “For curious researchers, the Place Pulse dataset can even be used to study the association between urban perception and other datasets, such as violent crime, creativity or economic growth.” In other words, does the perception that a place looks crime-ridden actually correlate with crime, and if so what policy decisions could be drawn from that?
At present, only city and trend data is being released, not individual neighborhood data, but the project’s principal investigator, César Hidalgo, hinted that in the future even more information might be made public once enough users click and rate.
Hidalgo was also behind MIT’s PRISM-like meta-data tool as well as the brainy online talk show, Cambridge Nights.