Two Spanish researchers have shown how Twitter data can be used to pinpoint places where people get together at certain times of the day for specific activities. Their approach provides a potentially valuable alternative to standard questionnaire-based surveys, as it also enables town planners to take account of prevalent night time activities in urban areas.

Twitter analysis of potential value for urban planners

What if the selfie you took with your iPhone in the latest trendy bar and then posted on Facebook, or the tweet published during a party, could actually serve the public good? If this notion seems a little strange, it is nevertheless based on a very simple observation. The huge quantity of content published every day on the social networks constitutes a massive pool of diverse information available for anyone and everyone to use. Now that smartphones and tablets have become so widespread, with an estimated 1.28 billion smartphones sold worldwide in 2014, the quantity of data has continued to rise, with an additional element – geolocation data. Recently Enrique and Vanessa Frias-Martinez, Spanish brother and sister researchers at Telefonica Research in Madrid and the University of Maryland, USA, respectively, came up with the idea of using this new factor to assist with urban planning. Their research, summarised in a paper entitled Crowdsourcing Land Use Maps via Twitter, published in the journal Engineering Applications of Artificial Intelligence, focuses mainly on Twitter, whose 500 million active accounts constitute an excellent freely available database of digital information.

Citizens’ use of urban land space                                                  

“For example, Twitter includes longitude-latitude information in the tweet if the user so desires. Among possible applications, we have seen that this network could be highly suited to helping in urban planning, especially in identifying land use,” explains Enrique Frias-Martinez in the conclusions reported on the Technobahn website. The two computer scientists have developed a technique which gives an automated reading of how population densities vary in different parts of a city according to what people are doing there. They centred their research on Manhattan (New York) and two European cities: Madrid and London. For Manhattan and Madrid they identified four different uses of land: residential, business, daytime leisure (mainly parks and tourist areas) and nightlife areas. In London, they also highlighted areas of industrial land use. The results were then validated with open data sources.

Incorporating nightlife into urban planning                               

Enrique Frias-Martinez believes that this technique could replace the traditional questionnaire-based survey approach to urban planning. Using Twitter, “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 (…) is very costly and can cause problems due to the lack of accuracy of the answers,” he points out. Frias-Martinez sees another specific advantage in his method: "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.” This innovative approach could also help local authorities to deploy their police forces and cleaning staff more efficiently, and to look at the overall impact bars and nightclubs have on residential areas. The method is of course not infallible. Geolocation is after all optional on Twitter, and not everyone systematically makes their comings and goings public.


This map shows which areas are used for business, leisure and nightlife in Madrid. The uncoloured part corresponds to residential areas. The study revealed that night tweeting activity is concentrated on weekends in Madrid, but on weekdays in Manhattan. (Source: V. and E. Frías-Martínez)

By Guillaume Renouard