We analyze a massive social network, gathered from the records of a large mobile phone operator, with more than a million users and tens of millions of calls. We examine the distributions of the number of phone calls per customer; the total talk minutes per customer; and the distinct number of calling partners per customer. We find that these distributions are skewed, and that they significantly deviate from what would be expected by power-law and lognormal distributions.
To analyze our observed distributions (of number of calls, distinct call partners, and total talk time), we propose PowerTrack , a method which fits a lesser known but more suitable distribution, namely the Double Pareto LogNormal (DPLN) distribution, to our data and track its parameters over time. Using PowerTrack , we find that our graph changes over time in a way consistent with a generative process that naturally results in the DPLN distributions we observe. Furthermore, we show that this generative process lends itself to a natural and appealing social wealth interpretation in the context of social networks such as ours. We discuss the application of those results to our model and to forecasting.
%0 Conference Paper
%1 DBLP:conf/kdd/SeshadriMSBFL08
%A Seshadri, Mukund
%A Machiraju, Sridhar
%A Sridharan, Ashwin
%A Bolot, Jean
%A Faloutsos, Christos
%A Leskovec, Jure
%B KDD
%D 2008
%K graph.theory network power-law social.network
%P 596-604
%R 10.1145/1401890.1401963
%T Mobile call graphs: beyond power-law and lognormal distributions
%X We analyze a massive social network, gathered from the records of a large mobile phone operator, with more than a million users and tens of millions of calls. We examine the distributions of the number of phone calls per customer; the total talk minutes per customer; and the distinct number of calling partners per customer. We find that these distributions are skewed, and that they significantly deviate from what would be expected by power-law and lognormal distributions.
To analyze our observed distributions (of number of calls, distinct call partners, and total talk time), we propose PowerTrack , a method which fits a lesser known but more suitable distribution, namely the Double Pareto LogNormal (DPLN) distribution, to our data and track its parameters over time. Using PowerTrack , we find that our graph changes over time in a way consistent with a generative process that naturally results in the DPLN distributions we observe. Furthermore, we show that this generative process lends itself to a natural and appealing social wealth interpretation in the context of social networks such as ours. We discuss the application of those results to our model and to forecasting.
@inproceedings{DBLP:conf/kdd/SeshadriMSBFL08,
abstract = {We analyze a massive social network, gathered from the records of a large mobile phone operator, with more than a million users and tens of millions of calls. We examine the distributions of the number of phone calls per customer; the total talk minutes per customer; and the distinct number of calling partners per customer. We find that these distributions are skewed, and that they significantly deviate from what would be expected by power-law and lognormal distributions.
To analyze our observed distributions (of number of calls, distinct call partners, and total talk time), we propose PowerTrack , a method which fits a lesser known but more suitable distribution, namely the Double Pareto LogNormal (DPLN) distribution, to our data and track its parameters over time. Using PowerTrack , we find that our graph changes over time in a way consistent with a generative process that naturally results in the DPLN distributions we observe. Furthermore, we show that this generative process lends itself to a natural and appealing social wealth interpretation in the context of social networks such as ours. We discuss the application of those results to our model and to forecasting.},
added-at = {2011-03-29T00:46:02.000+0200},
author = {Seshadri, Mukund and Machiraju, Sridhar and Sridharan, Ashwin and Bolot, Jean and Faloutsos, Christos and Leskovec, Jure},
bibsource = {DBLP, http://dblp.uni-trier.de},
biburl = {https://www.bibsonomy.org/bibtex/21fba1ebb059cb09774cf3edc23ee5bb6/ytyoun},
booktitle = {KDD},
crossref = {DBLP:conf/kdd/2008},
doi = {10.1145/1401890.1401963},
interhash = {673d54c72f19a51f10d0c9748c0499cf},
intrahash = {1fba1ebb059cb09774cf3edc23ee5bb6},
keywords = {graph.theory network power-law social.network},
pages = {596-604},
timestamp = {2017-05-29T13:54:56.000+0200},
title = {Mobile call graphs: beyond power-law and lognormal distributions},
year = 2008
}