Discovering and Understanding City Events with Big Data: The Case of Rome

The increasing availability of large amounts of data and digital footprints has given rise to ambitious research challenges in many fields, which spans from medical research, financial and commercial world, to people and environmental monitoring. Whereas traditional data sources and census fail in capturing actual and up-to- date behaviors, Big Data integrate the missing knowledge providing useful and hidden information to analysts and decision makers. In the domain of city and human dynamics, with this paper we have a twofold objective: on the one hand we want to detect and characterize important and unusual events at urban level; to the other hand, we want to estimate the composition of the population who attended to them. The goal is to identify and isolate significant and unusual peaks of presences among all the mobile signals of presences registered during the days, thus quantifying the events and their impact on the city dynamics and population composition. With respect to state of the art, this work exploit a more precise identification and characterization of the events of the different city users through the categorization of phone users . Such an estimation is performed through the Sociometer, a tool for urban demographics which processes mobile phone records to characterize city users into behavioral categories (figure 1). The former proposes a method to detect unusual events relying on users’ mobility profile, considering each antenna the user connected to as a location. Then, they identify unusual crowds detecting users whom are aggregating in areas unusual for them, according to their corresponding mobility profiles (figure 2 and 3); the latter proposes a supervised approach to learn the pattern of an event, that differs from our method since it is based on the availability of a list of known events. The result we obtain is particularly useful to support the understanding of phenomena and provide new knowledge to decision makers and urban planners. For example, it was possible to eliminate noise signals introduced by the so called “people in transit”. Moreover, this distinction allowed us to focus on the impact of the events on the city users identifying what kind of people are interested in different kind of events, and try to answer to typical questions like: does the event attract people from outside the city? Or, is it rather attended especially from residents? How does the city users composition change during these events, w.r.t. the normal days? Our aim is to provide updated knowledge in order to improve demographic and urban investigations and support decision makers in planning purposes.