San Francisco, CA, United States
San Francisco, CA, United States
SEARCH FILTERS
Time filter
Source Type

Leo Y.,University of Lyon | Fleury E.,University of Lyon | Alvarez-Hamelin J.I.,University of Buenos Aires | Sarraute C.,GranData | Karsai M.,University of Lyon
Journal of the Royal Society Interface | Year: 2016

The uneven distribution of wealth and individual economic capacities are among the main forces, which shape modern societies and arguably bias the emerging social structures. However, the study of correlations between the social network and economic status of individuals is difficult due to the lack of large-scale multimodal data disclosing both the social ties and economic indicators of the same population. Here,we close this gap through the analysis of coupled datasets recording the mobile phone communications and bank transaction history of one million anonymized individuals living in a Latin American country. We show that wealth and debt are unevenly distributed among people in agreementwith the Pareto principle; the observed social structure is strongly stratified, with people being better connected to others of their own socioeconomic class rather than to others of different classes; the social network appears to have assortative socioeconomic correlations and tightly connected 'rich clubs'; and that individuals from the same class live closer to each other but commute further if they are wealthier. These results are based on a representative, society-large population, and empirically demonstrate some long-lasting hypotheses on socioeconomic correlations, which potentially lay behind social segregation, and induce differences in human mobility. © 2016 The Author(s) Published by the Royal Society. All rights reserved.


Oskarsdottir M.,Catholic University of Leuven | Bravo C.,University of Southampton | Verbeke W.,Vrije Universiteit Brussel | Sarraute C.,GranData | And 3 more authors.
Expert Systems with Applications | Year: 2017

Social network analytics methods are being used in the telecommunication industry to predict customer churn with great success. In particular it has been shown that relational learners adapted to this specific problem enhance the performance of predictive models. In the current study we benchmark different strategies for constructing a relational learner by applying them to a total of eight distinct call-detail record datasets, originating from telecommunication organizations across the world. We statistically evaluate the effect of relational classifiers and collective inference methods on the predictive power of relational learners, as well as the performance of models where relational learners are combined with traditional methods of predicting customer churn in the telecommunication industry. Finally we investigate the effect of network construction on model performance; our findings imply that the definition of edges and weights in the network does have an impact on the results of the predictive models. As a result of the study, the best configuration is a non-relational learner enriched with network variables, without collective inference, using binary weights and undirected networks. In addition, we provide guidelines on how to apply social networks analytics for churn prediction in the telecommunication industry in an optimal way, ranging from network architecture to model building and evaluation. © 2017 Elsevier Ltd


Luo S.,City College of New York | Morone F.,City College of New York | Sarraute C.,GranData | Travizano M.,GranData | Makse H.A.,City College of New York
Nature Communications | Year: 2017

It is commonly believed that patterns of social ties affect individuals' economic status. Here we translate this concept into an operational definition at the network level, which allows us to infer the economic well-being of individuals through a measure of their location and influence in the social network. We analyse two large-scale sources: telecommunications and financial data of a whole country's population. Our results show that an individual's location, measured as the optimal collective influence to the structural integrity of the social network, is highly correlated with personal economic status. The observed social network patterns of influence mimic the patterns of economic inequality. For pragmatic use and validation, we carry out a marketing campaign that shows a threefold increase in response rate by targeting individuals identified by our social network metrics as compared to random targeting. Our strategy can also be useful in maximizing the effects of large-scale economic stimulus policies. © 2017 The Author(s).


Chen G.,French Institute for Research in Computer Science and Automation | Viana A.C.,French Institute for Research in Computer Science and Automation | Sarraute C.,GranData
2017 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2017 | Year: 2017

Call Detail Records (CDRs) are a primary source of whereabouts in the study of multiple mobility-related aspects. However, the spatiotemporal sparsity of CDRs often limits their utility in terms of the dependability of results. In this paper, driven by real-world data across a large population, we propose two approaches for completing CDRs adaptively, to reduce the sparsity and mitigate the problems the latter raises. Owing to high-precision sampling, the comparative evaluation shows that our approaches outperform the legacy solution in the literature in terms of the combination of accuracy and temporal coverage. Also, we reveal those important factors for completing sparse CDR data, which sheds lights on the design of similar approaches. © 2017 IEEE.


Leo Y.,University of Lyon | Busson A.,University of Lyon | Sarraute C.,GranData | Fleury E.,University of Lyon
Ad Hoc Networks | Year: 2016

In urban areas, the population density is still growing (the population density starts exceeding 20.000 inhabitants per km2), and so, the density of mobile users becomes very important. People are moving from home to work, from work to active places. One can take benefit of the mobility and the density to justify DTN (Delay Tolerant Network) approach protocol to convey SMS (or alternative messaging services) traffic. Indeed, the mobility of users, especially during the day, create an ad hoc mobile network where the nodes are the smartphones hold by mobile clients. In this paper, their performance evaluations are based on a measurement and analysis of SMS traces coming from a nationwide cellular telecommunication operator during a two month period, we propose several DTN like basic network protocols for delivering SMS. We perform a temporal and spatial analysis of the Mexico City cellular network considering geolocalized SMS to characterize the traffic. Such key characterization allows us to answer the question: is it possible to transmit SMS using phones as relay in a large city such as Mexico City? We define four network protocols to transmit SMS from a source to a destination. We study a mobile dataset including 8 Million users living in Mexico city. This gives us a precise estimation of the average transmission time and the global performance of our approach. Our analysis shows that after 30 min, half of the SMS are delivered successfully to destination. On the contrary to the cellular networks, we explain how much the potentiality of the mobile users network can take benefit from complementary properties such as the locality of SMS, the density of phones in Mexico City and the mobility of phone users. Moreover, we show that in a realistic scenario, our approach induces reasonable storage cost. © 2016 Elsevier B.V.


Leo Y.,University of Lyon | Busson A.,University of Lyon | Sarraute C.,GranData | Fleury E.,University of Lyon
Computer Communications | Year: 2016

Cellular technologies are evolving quickly to constantly adapt to new usage and tolerate the load induced by the increasing number of phone applications. Understanding the mobile traffic is thus crucial to refine models and improve experiments. In this context, one has to understand the temporal activity of a user and the user movements. At the user scale, the usage is not only defined by the amount of calls but also by the user's mobility. At a higher level, the base stations have a key role on the quality of service. In this paper, we analyze a very large Call Detail Records (CDR) over 12 months in Mexico. It contains 8 millions users and 5 billions of call events. Our first contribution is the study call duration and inter-arrival time parameters. Then, we assess user movements between consecutive calls (switching from a station to another one). Our study suggests that user mobility is pretty dependent on user activity. Furthermore, we show properties of the inter-call mobility by making an analysis of the call distribution. © 2016 Elsevier B.V.


PubMed | GranData, University of Buenos Aires and University of Lyon
Type: Journal Article | Journal: Journal of the Royal Society, Interface | Year: 2016

The uneven distribution of wealth and individual economic capacities are among the main forces, which shape modern societies and arguably bias the emerging social structures. However, the study of correlations between the social network and economic status of individuals is difficult due to the lack of large-scale multimodal data disclosing both the social ties and economic indicators of the same population. Here, we close this gap through the analysis of coupled datasets recording the mobile phone communications and bank transaction history of one million anonymized individuals living in a Latin American country. We show that wealth and debt are unevenly distributed among people in agreement with the Pareto principle; the observed social structure is strongly stratified, with people being better connected to others of their own socioeconomic class rather than to others of different classes; the social network appears to have assortative socioeconomic correlations and tightly connected rich clubs; and that individuals from the same class live closer to each other but commute further if they are wealthier. These results are based on a representative, society-large population, and empirically demonstrate some long-lasting hypotheses on socioeconomic correlations, which potentially lay behind social segregation, and induce differences in human mobility.


Sarraute C.,GranData | Blanc P.,University of Buenos Aires | Burroni J.,GranData
ASONAM 2014 - Proceedings of the 2014 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining | Year: 2014

Mobile phone usage provides a wealth of information, which can be used to better understand the demographic structure of a population. In this paper we focus on the population of Mexican mobile phone users. Our first contribution is an observational study of mobile phone usage according to gender and age groups. We were able to detect significant differences in phone usage among different subgroups of the population. Our second contribution is to provide a novel methodology to predict demographic features (namely age and gender) of unlabeled users by leveraging individual calling patterns, as well as the structure of the communication graph. We provide details of the methodology and show experimental results on a real world dataset that involves millions of users. © 2014 IEEE.


Ponieman N.B.,GranData | Salles A.,University of Buenos Aires | Sarraute C.,GranData
Proceedings of the 2013 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2013 | Year: 2013

The massive amounts of geolocation data collected from mobile phone records has sparked an ongoing effort to understand and predict the mobility patterns of human beings. In this work, we study the extent to which social phenomena are reflected in mobile phone data, focusing in particular in the cases of urban commute and major sports events. We illustrate how these events are reflected in the data, and show how information about the events can be used to improve predictability in a simple model for a mobile phone user's location. Copyright 2013 ACM.


Loading GranData collaborators
Loading GranData collaborators