Current data collection can be expensive and labor-intensive. Some researchers think social media data mining can help.
Social media platforms like Facebook and Twitter generate tons of user data, which has been used for marketing by companies and organizations. Now a couple of researchers from Carnegie Mellon University see potential in mining that data for better urban planning. Dan Tasse and Jason Hong write that current data collection, like the census, can be a very expensive and labor-intensive undertaking while data from social media is quick, public, and relatively easy to access. “We believe that this kind of geotagged social media data, combined with new kinds of analytics tools, will let urban planners, policy analysts, social scientists, and computer scientists explore how people actually use a city, in a manner that is cheap, highly scalable, and insightful,” they write in their paper.
For example, Tasse and Hong write that the data could potentially be used to map socioeconomic factors, and to track various quality of life factors – violence, transportation availability, grocery store usage – within a city or neighborhood. With that data in hand, they write, decision makers might better understand not just who lives in a city and how many restaurants are on a particular block, but who interacts with what businesses and how, too, which could lead to better planning and a re-evaluation of the needs of a community. And since social media data monsoons from our devices continuously – about 500 million tweets per day – that re-evaluation could happen much more frequently than the decennial census data allows.
This is something the Urban Attitudes Lab at Tufts University is also interested in. Other researchers in the U.S. and beyond are examining ways to use social media data to track people’s migration, or to look at urban nightlife activity. Maybe in the future transportation officials can decide on the number of buses running a route that includes a main entertainment drag based on Twitter activity there on a particular night. Or perhaps new bike path routes between neighborhoods will take into account data on cycling traffic generated from social media.
Sounds like urban planning science fiction, right? The use of such data isn’t without problems, though. For one, social media users are self-selective; young, male, tech-savvy people tend to use it more. And their use of the medium can be performative. One might readily check in at a hip restaurant, but not, as Tasse and Hong point out, at a fast food establishment. After all, data is a side effect of social media use and not the goal of it for users. As such, the data may not be representative of a city and its residents, but rather of very specific groups of residents. “For now, one approach is to look for ways of accounting for these kinds of biases in models,” Tasse and Hong write. “Another approach is to make clearer what the models do and do not represent.” So there’s still a lot of work left, but the future of urban planning looks data-rich.