Location relevance and diversity in symbolic trajectories with application to telco data

Abstract

We present an approach to the discovery and characterization of relevant locations and related mobility patterns in symbolic trajectories built on call detail records-CDRs-of mobile phones (telco trajectories). While the discovery of relevant locations has been widely investigated for continuous spatial trajectories (e.g., stay points detection methods), it is not clear how to deal with the problem when the movement is defined over a discrete space and the locations are symbolic, noisy and irregularly sampled, such as in telco trajectories. In this paper, we propose a methodological approach structured in two steps, called trajectory summarization and summary trajectories analysis, respectively, the former for removing noise and irrelevant locations; the latter to synthesize key mobility features in a few novel indicators. We evaluate the methodology over a dataset of approx 17,000 trajectories with 55 million points and spanning a period of 67 days. We find that trajectory summarization does not compromise data utility, while significantly reducing data size. Moreover, the mobility indicators provide novel insights into human mobility behavior.

On Location Relevance and Diversity in Human Mobility Data

“…Summary of contributions. This article significantly extends the preliminary version in [15] to address a few open questions, focusing in particular on the consolidation, enrichment and validation of the analytical framework, built on the trajectory summarization technique. In addition, we report a richer set of experiments on a significantly larger dataset of telco trajectories.…”

“…in [1]. In addition, we test the analytical framework on a dataset of anonymized 100,000+ telco trajectories (superset of the dataset used in [15]). Although larger datasets of telco trajectories are reported in literature, e.g.…”

Learning Behavioral Representations of Human Mobility

“…Core resources for the study of human mobility are the datasets of Call Detail Records (CDR) [5]. CDRs report the communication activities of mobile communication subscribers as series of geo-referenced events, i.e., voice call start/end, text message, data upload/download, collected by mobile operators for billing purposes [9]. We refer to a series of temporally ordered CDRs associated to an individual as CDR trajectory.…”

“…High values of R g are interpreted as high user's propensity to mobility. Another kind of measure is true location diversity, an entropy-based metric that we have recently proposed to quantify the heterogeneity of locations in symbolic trajectories [9]. In general, however, the approaches grounded on the choice of metrics present an important shortcoming, namely it is not clear which mobility features are really significant to characterize the mobility behavior.…”

“…• It leverages a technique we have recently proposed for the removal of noise and extraction of significant locations from CDR and symbolic trajectories, called SeqScan-d. The resulting trajectories are called summary trajectories [9].…”