Capturing movements and opinions in urban open spaces through the analysis of user generated data

Elena Masala
Jan 12th – Febr 20th 2015

Focus of the author’s STSM in Barcelona is the analysis of possible improvements in the design of open urban spaces by means of user generated data coming from social network platforms and apps. To carrying on the study, two sets of data have been used: the first one is given by the gathering of Tweets sent within the metropolitan area of Barcelona during the period January 7-19, 2015. The second set is the collection of records generated by the CYBERPARKS App during its pilot test in Barcelona in the afternoon of November 27, 2014.

The experimental study within the STSM has been organised in different steps. First step concerned the analysis of the status of the art regarding the application of user generated content data within spatial studies. Second step consisted in setting the methodology to deal with the objectives of the STSM and
its application to the two case studies in Barcelona. The final step has provided the conclusions derived from the direct experience of STSM with a particular reasoning of possible future developments for the integration of new methodologies within spatial planning processes.

The analysis status of the art highlighted a diffuse and a worldwide interest in the use of both Open and Big data, especially in their possible applications for the improvement of quality of life in cities. The analysis of data coming from the two platforms and their comparison provided outcomes on the possible use of such data within the planning and design of public open spaces. This study pointed out also some guideline on possible applications of user generated content (UGC) data to collect information that could be useful and effective in the design processes. In particular, the research produced a number of possible maps which can be obtained by the analysis and elaboration of Twitter data. The first map generated by tweets data shows the density of message sent per walkable cell. A second map connects the tweets sent by each single user according to their temporal sequence. The same visualisation was realised considering only the users who tweeted from almost one of the two case study areas. A further map was generated using the subsequent Tweets location overlaid on the road network downloaded from the Open Street Map (OSM) portal (Open Street Map Community, 2004). Using the shortest-path algorithm and the actual road system, the width of road line express the density of Tweets on roads. Splitting the Tweets according their time attribute, other maps show variations in the use of urban space during the week days and along the 24 daily hours. Finally, mapping the users’ paths according to the language setting in users’ profile allows to generate maps on the use of the city made by local or tourist people.

To conclude, an overview of pros and cons in the use of UGC data provides considerations on possible effective uses, while a list for future developments is illustrated to show how the research can evolve in the next months.