Urban Pulse maps, analyzes use of urban spaces
The amount of data collected by public agencies in many cities is staggering, including such granular information as the time and location of taxicab pickups and drop-offs and data on entry and exits of passengers using the subway. The amount of location data being collected by social media such as Twitter and Flickr -- not to mention location data being collected by cell phone service providers -- is similarly massive.
Fabio Miranda, a Ph.D. candidate at New York University’s Tandon School of Engineering, has developed an open-source tool that allows city planners to combine and visualize such datasets over time. Dubbed Urban Pulse, the software makes it possible to visualize how, when and -- to a degree -- by whom urban spaces are being used.
“We have this explosion of open datasets available that describe how the city behaves,” Miranda said, “so we were looking for how we could use these datasets to make sense of the city.”
The first iteration offered by Miranda and his colleagues employs data sets from Flickr and Twitter. The Flickr data is used to track tourist movement, and the Twitter data is treated by Urban Pulse as a measure of population densities at given locations. (Yes, non-tourists use Flickr too, but the team believes the preponderance of use is by tourists.) Areas with the most Flickr and Twitter activity are displayed as “hot pulses.” The displayed data can also be analyzed through any specified period of time to see the ebb and flow of activity in locations throughout the city over time.
“It's purely a data-driven approach,” Miranda said. “You can explore how a given region in the city compares with another region in another city or within the same city. You can see how the activity changes throughout time and throughout different resolutions of time, and then that can give you new insights as to how that public space has been used.”
While Urban Pulse currently only uses location data from Twitter and Flicker, Miranda said those who download the software from GitHub are free to upload other datasets. For his part, Miranda said he is looking to add in data on taxi pickups and drop-offs as well as subway activity. Those datasets, he says, are publicly available in New York City.
Miranda also is planning to apply the same technology to analyzing the impact of shadows on different parts of a city through time. “Let's say you have Central Park and you want to see how the buildings around Central Park impact the use of public spaces at, say, 3 p.m. in each day or over an entire year,” he said. He added that the data could also be helpful in, for example, understanding the impacts on flora in the park.
Posted by Patrick Marshall on Nov 08, 2017 at 12:21 PM