Identifying user habits through data mining on call data records

Abstract

In this paper we propose a framework for identifying patterns and regularities in the pseudoanonymized Call Data Records (CDR) pertaining a generic subscriber of a mobile operator. We face the challenging task of automatically deriving meaningful information from the available data, by using an unsupervised procedure of cluster analysis and without including in the model any apriori knowledge on the applicative context. Clusters mining results are employed for understanding users' habits and to draw their characterizing profiles. We propose two implementations of the data mining procedure; the first is based on a novel system for clusters and knowledge discovery called LD-ABCD, capable of retrieving clusters and, at the same time, to automatically discover for each returned cluster the most appropriate dissimilarity measure (local metric). The second approach instead is based on PROCLUS, the well-know subclustering algorithm. The dataset under analysis contains records characterized only by few features and, consequently, we show how to generate additional fields which describe implicit information hidden in data. Finally, we propose an effective graphical representation of the results of the data-mining procedure, which can be easily understood and employed by analysts for practical applications.

Approaches, technologies and importance of analysis of the number of train travellers

“…• Internet traffic, including activity on social media and data from search engines • Movement-related data, and different visualizations, including pictures, video, BIM models and maps • Physical environment, typically from different types of sensors • Commercial activity, the use of payment services and consumption patterns New types of data open many possibilities for the analysis of transport measures (Barabino et al, 2014;Bianchi, Rizzi, Sadeghian, & Moiso, 2016). A significant portion of Big Data is geospatial data, generated from sources such as mobile devices and RFID sensors.…”

Use of mobile phone data for analysis of number of train travellers

“…CDRs are generated by phone communication activities and contain relevant information about the activity (e.g., caller/callee, time, duration) and the location of the cell phone tower that handles the communication (Zhao et al, 2016). Studies have shown that CDR data can be used to study habits and mobility patterns of mobile users (Bianchi et al, 2016;Zhao et al, 2016), to study user movements (Leo et al, 2016), and to calculate commuting matrices with a very high level of accuracy (Frias-Martinez, et al, 2012). Studies have also looked at utilizing mobile data to estimate intra-city travel time (Kujala, Aledavood, & Saramäki, 2016) and have shown that mobile data could be employed as a real-time traffic monitoring tool (Järv et al, 2012).…”

“…With pseudo-anonymised data (i.e., the ID is replaced with a code), the record must be pre-processed to reduce probability of re-identification. A common procedure is to decrease time resolution or increase space granularity (Bianchi et al, 2016). Norwegian Law states that collected personal information should only be used for the specific purpose for which it was originally collected (Drageide, 2009).…”

Sources of VGI for Mapping

“…These data have been analysed to investigate applications in areas such as transportation planning (Di Lorenzo et al, 2016), user behaviour (Bianchi et al, 2016), public health (Oliver et al, 2015), the spatial spread of diseases such as cholera (Bengtsson et al, 2015) or population displacement after a major disaster (Wilson et al, 2016).…”