Abstract
We present a methodology to automatically identify users’ relevant places from cellular network data.11The approach described in this paper is the subject of Patent Application PCT/EP2014/058003 filed by Telecom Italia on April 2014. In this work we used anonymized Call Detail Record (CDR) comprising information on where and when users access the cellular network. The key idea is to effectively cluster CDRs together and to weigh clusters to determine those associated to frequented places. The approach can identify users’ home and work locations as well as other places (e.g., associated to leisure and night life). We evaluated our approach threefold: (i) on the basis of groundtruth information coming from a fraction of users whose relevant places were known, (ii) by comparing the resulting number of inhabitants of a given city with the number of inhabitants as extracted by the national census. (iii) Via stability analysis to verify the consistency of the extracted results across multiple time periods. Results show the effectiveness of our approach with an average 90% precision and recall.
Highlights
The widespread diffusion of mobile phones and cellular networks provides a practical way to collect location-based information from a large user population
In the following of this Section we present the key elements of the proposed approach: (i) we provide some details on the Call Detail Record (CDR) dataset on the basis of our proposal. (ii) We describe the function we use to assign a weight to each CDR.. (iii) We discuss the approach we used to cluster CDRs in well-defined areas. (iv) We present our mechanism to assign an importance weight to clusters. (v) we describe an automatic thresholding approach to identify those clusters associated to relevant places
Mobility data is obtained from Call Detail Records (CDR) and Mobility Management (MM) procedure messages
Summary
The widespread diffusion of mobile phones and cellular networks provides a practical way to collect location-based information from a large user population. We propose an innovative approach to automatically identify the places that people routinely frequent (e.g., home, work place, favorite nightlife locations) from the analysis of anonymized positioning data from a cellular network (i.e., CDR—Call Detail Record). The knowledge of such relevant places finds important applications in mobile services, marketing, traffic forecasting, urban planning and management services [1,2,3,4] (see Section 2 for more details). Our proposal improves the state of the art in all the three phases: clustering, weighing, thresholding
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.