Enrique and Vanessa Frías-Martínez, brother and sister computer science researchers at Telefonica Research and the University of Maryland, looked at tweet aggregates from Manhattan, Madrid and London, with their findings published in the journal Engineering Application of Artificial Intelligence.
In Manhattan and Madrid, the researchers identified four land uses: residential, business, daytime leisure (mainly parks and tourist areas) and nightlife areas. In London, they also established industrial land uses. The results were validated with open data sources.
The siblings put together maps showing how some of the urban land was used after looking at "a clustering of geographical regions with similar tweeting activity patterns."
The study found that in Madrid, night tweeting activity is concentrated on weekends, and in Manhattan, the same could be said for weekdays.
On the other hand, London is characterized by its tweeting activity in daytime leisure.
"One of the most interesting contributions of the study is the identification of nightlife areas, since this type of land use in not often specified in urban planning, despite the problems of noise, security and need for cleaning that this creates," Enrique Frias-Martinez said.
"A forgotten issue in urbanism is land use during the night time," and problems unique to those hours could be improved with geotagging research, he said.
Faster, more accurate than surveys
People generate vast amounts of geolocalized content when they post information to social media websites — an activity that's becoming widespread through the use of cellphone apps.
Geotagging technology inside smartphones is usually automatically set to "on," meaning photos captured on the phone and sent to social media will include metadata that gives away the user's location, unless the feature is disabled.
Twitter allows people to add a location to individual tweets in the compose box. If you forget to turn the feature off, your location will be automatically revealed the next time you tweet.
Using Twitter, says Enrique Frías-Martínez, "you can capture information on urban land use more efficiently and for a much larger number of people than with questionnaires. Moreover, this type of consultation, traditionally used until now in planning activities, [is] very costly and can cause problems due to the lack of accuracy of the answers."