Mobile devices are today noticeable means by which people disseminate digital tracks of their everyday activities: movements, purchase transactions, preferences, opinions, and so on. In particular, mobile phones and the data they produce revealed as a high-quality proxy for studying people mobility in different domains, such as environmental monitoring, transportation planning , smart cities and social relationship analysis and innovative demographic indexes.
Currently, a hot topic in the modernization of official statistics is precisely how to use big data in combination with traditional data sources, in order to improve quality, timeliness and spatio-temporal granularity of statistical information. Moreover, new technologies allow scientific communities to expand the horizon of the data analysis to provide a real and up-to-date support to decision makers.
KDDLab (ISTI - CNR, Pisa) developed a methodology able to classify city users into behavioral categories (e.g. residents, Commuter, Visitors) using their mobile phone traces called Sociometer. It was tested by several extensive case study, demonstrating how it improves the discovery and understanding of new hidden patterns and users behaviors. The analytical process is based on a synthesized representation of the call patterns, named Individual Call Profile, which summarizes the call activities of a user during weekdays and weekends at three different time slots (early morning, work hours, and evenings). Using the state-of-art technologies such as Hadoop-Spark, the Sociometer is able to process big quantity of mobile phone data.
Sociometer is able to quantify the number of real presences in a given territories in order to measure in effective way the real residents and the real impact of the events within cities. Sociometer, furthermore, identifies the contribution of different categories of people to the events and to establish the origins of the attendees. The collaboration and feedbacks of the Local Administration were decisive for the refinement and evaluation of the process. Ongoing collaboration among KDDLab and National Statistic Offices (i.e Istat), has the objective to use Sociometer to extract new demographic indexes.
1. B. Furletti, L. Gabrielli, C. Renso, and S. Rinzivillo. 2013. Analysis of GSM calls data for understanding user mobility behavior. In Proc. of IEEE Big Data. 550-555.
2. B. Furletti, L. Gabrielli, R. Trasarti, F. Giannotti, and D. Pedreschi. 2015. City users' classification with mobile phone data. In Proc. of IEEE Big Data. 1007-1012.
3. B. Furletti, L. Gabrielli, G. Garofalo, F. Giannotti, L. Milli, M. Nanni, D. Pedreschi, and R. Vivio. 2014. Use of mobile phone data to estimate mobility flows. Measuring urban population and inter-city mobility using big data in an integrated approach. In Proc. of SIS 2014).
City users' classification with mobile phone data: the Sociometer