New York City is famously divided into five boroughs, each with its own quirks and cultural nuances. So much of how the city is understood comes from the political boundaries of Manhattan, the Bronx, Queens, Brooklyn, and Staten Island, the boundaries of which were drawn up in the 1800s. Topos, an urbanism-oriented artificial intelligence startup, thinks it’s a woefully obsolete and outdated way to think about the city.
The boroughs originally existed independently, each with its own governance structure, but they consolidated in the late 1800s to bolster the region’s economy. Their borders were roughly defined by waterways, but today the city is interconnected by bridges, tunnels, and subway lines, and borough-level governments wield little political power compared to the city council. For those reasons, Topos redesigned the city’s five boroughs based on how similar they are in neighborhood composition. The goal was to get a better understanding of how the city is really organized.
Topos analyzed the city at zip-code level with an algorithm that used satellite images, topographic data, ambient light level information, building information, and other data to group areas with similar characteristics. The program narrowed down all of the variables that account for neighborhood differences, to the handful that account for the most dramatic distinctions between them: population density, land use, building type, and business concentration.
What emerged were five different clusters. Minihattan, as Topos named the first one, is a “heavily built-up vertical borough” that is contained in Manhattan (green on the map). The next is called “the Ring,” which has expensive residential buildings and lots of different business types but is shorter than Minihattan; this distinction includes the Upper West and Upper East sides of Manhattan and parts of Harlem, northern Brooklyn neighborhoods, and the Queens neighborhoods of Forest Hills, Astoria, and Long Island City.
Meanwhile, about half of New York City’s residents live in “North and South Bend,” predominately residential neighborhoods with less commercial variation than the Ring or Minihattan. These are denoted in blue and red on the map and include Washington Heights, the Bronx, southern Brooklyn, northeastern Staten Island, and central Queens. The fifth borough, as Topos sees it, is called “the Meadows” (denoted in yellow) and is characterized by leafy, low-density suburban homes. This includes neighborhoods along the far eastern shores of the Bronx, Staten Island, eastern Queens, and the Rockaways.
Topos believes that artificial intelligence will reshape how cities are understood and could potentially alter how planners and policy makers engage with urban areas.
“AI is essential for smart urban planning because it can give us a
real-time understanding of what cities feel like moment to moment
rather than decade to decade,” Topos cofounder Will Shapiro tells Co.Design via email. “Unlike manual tools such as the U.S. Census, our platform enables a dynamic, highly granular, and globally scalable view of cities and neighborhoods. By pulling from a wide variety of technologies and disciplines–computer vision, natural language processing, network science, machine learning, statistics, topology, urbanism, data visualization, and information design–we are going beyond more familiar demographic viewpoints to capture the personality of a place, and what it feels like to actually be there.”
See the maps in the slide show above and read more about the process on Medium.