Link-centric analysis of variation by demographics in mobile phone communication patterns

We present a link-centric approach to study variation in the mobile phone communication patterns of individuals. Unlike most previous research on call detail records that focused on the variation of phone usage across individual users, we examine how the calling and texting patterns obtained from call detail records vary among pairs of users and how these patterns are affected by the nature of relationships between users. To demonstrate this link-centric perspective, we extract factors that contribute to the variation in the mobile phone communication patterns and predict demographics-related quantities for pairs of users. The time of day and the channel of communication (calls or texts) are found to explain most of the variance among pairs that frequently call each other. Furthermore, we find that this variation can be used to predict the relationship between the pairs of users, as inferred from their age and gender, as well as the age of the younger user in a pair. From the classifier performance across different age and gender groups as well as the inherent class overlap suggested by the estimate of the bounds of the Bayes error, we gain insights into the similarity and differences of communication patterns across different relationships.

[1]  Nitesh V. Chawla,et al.  Inferring user demographics and social strategies in mobile social networks , 2014, KDD.

[2]  Sune Lehmann,et al.  Modeling the Temporal Nature of Human Behavior for Demographics Prediction , 2015, ECML/PKDD.

[3]  Kimmo Kaski,et al.  Communication with Family and Friends across the Life Course , 2015, PloS one.

[4]  Cai Lian Tam,et al.  Mobile Phone Usage Preferences: The Contributing Factors of Personality, Social Anxiety and Loneliness , 2014 .

[5]  Manuel Cebrián,et al.  Limited communication capacity unveils strategies for human interaction , 2013, Scientific Reports.

[6]  Peter E. Hart,et al.  Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.

[7]  Alex 'Sandy' Pentland,et al.  Erratum to: Improving official statistics in emerging markets using machine learning and mobile phone data , 2017, EPJ Data Science.

[8]  Carlos Sarraute,et al.  Inference of demographic attributes based on mobile phone usage patterns and social network topology , 2015, Social Network Analysis and Mining.

[9]  John Platt,et al.  Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods , 1999 .

[10]  Ronald E. Rice,et al.  Configurations of Relationships in Different Media: FtF, Email, Instant Messenger, Mobile Phone, and SMS , 2007, J. Comput. Mediat. Commun..

[11]  Michalis Faloutsos,et al.  Inferring cellular user demographic information using homophily on call graphs , 2013, 2013 Proceedings IEEE INFOCOM.

[12]  Joshua Evan Blumenstock,et al.  Differences in phone use between men and women: quantitative evidence from Rwanda , 2012, ICTD '12.

[13]  Jari Saramäki,et al.  Daily Rhythms in Mobile Telephone Communication , 2015, PloS one.

[14]  Kimmo Kaski,et al.  Tracking urban human activity from mobile phone calling patterns , 2017, PLoS Comput. Biol..

[15]  Pedro J. Zufiria,et al.  Prediction of Telephone User Attributes Based on Network Neighborhood Information , 2012, MLDM.

[16]  Mehmet Barış Horzum,et al.  Is problematic mobile phone use explained by chronotype and personality? , 2016, Chronobiology international.

[17]  Matteo Migheli.  The sibling effect on the consumption of phone services , 2016 .

[18]  Kimmo Kaski,et al.  Uncovering intimate and casual relationships from mobile phone communication , 2018, ArXiv.

[19]  Kagan Tumer,et al.  Estimating the Bayes error rate through classifier combining , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[20]  Didier Sornette,et al.  Discrete hierarchical organization of social group sizes , 2004, Proceedings of the Royal Society B: Biological Sciences.

[21]  Kimmo Kaski,et al.  Different patterns of social closeness observed in mobile phone communication , 2018, Journal of Computational Social Science.

[22]  Vanessa Frías-Martínez,et al.  A Gender-Centric Analysis of Calling Behavior in a Developing Economy Using Call Detail Records , 2010, AAAI Spring Symposium: Artificial Intelligence for Development.

[23]  James G. Phillips,et al.  Personality and self reported mobile phone use , 2008, Comput. Hum. Behav..

[24]  I. Jolliffe.  Principal Component Analysis , 2002 .

[25]  Robin I. M. Dunbar,et al.  Romance and reproduction are socially costly. , 2015 .

[26]  Carlos Sarraute,et al.  A study of age and gender seen through mobile phone usage patterns in Mexico , 2014, 2014 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2014).

[27]  Ian T. Jolliffe,et al.  Principal Component Analysis , 2002, International Encyclopedia of Statistical Science.

[28]  P. Seabright,et al.  Do women have longer conversations? Telephone evidence of gendered communication strategies , 2011 .

[29]  Kimmo Kaski,et al.  Quantifying gender preferences in human social interactions using a large cellphone dataset , 2019, EPJ Data Science.

[30]  Kimmo Kaski,et al.  Absence makes the heart grow fonder: social compensation when failure to interact risks weakening a relationship , 2016, EPJ Data Science.

[31]  Albert-László Barabási,et al.  Sex differences in intimate relationships , 2012, Scientific Reports.

[32]  Kadan Aljoumaa,et al.  Predicting customer’s gender and age depending on mobile phone data , 2019, Journal of Big Data.

[33]  Kimmo Kaski,et al.  Sex differences in social focus across the life cycle in humans , 2015, Royal Society Open Science.

[34]  Z. Smoreda,et al.  Gender-specific use of the domestic telephone , 2000 .

[35]  Jari Saramäki,et al.  Spatial patterns of close relationships across the lifespan , 2014, Scientific Reports.

[36]  Alex Pentland,et al.  Predicting Personality Using Novel Mobile Phone-Based Metrics , 2013, SBP.

[37]  Vincent D. Blondel,et al.  A survey of results on mobile phone datasets analysis , 2015, EPJ Data Science.

[38]  Jari Saramäki,et al.  Temporal motifs reveal homophily, gender-specific patterns, and group talk in call sequences , 2013, Proceedings of the National Academy of Sciences.

[39]  Adam D. I. Kramer,et al.  Detecting Emotional Contagion in Massive Social Networks , 2014, PloS one.

[40]  Kimmo Kaski,et al.  Social physics: uncovering human behaviour from communication , 2018, Advances in Physics: X.

[41]  Thomas V. Pollet,et al.  Exploring variation in active network size: Constraints and ego characteristics , 2009, Soc. Networks.

[42]  Alex 'Sandy' Pentland,et al.  Improving official statistics in emerging markets using machine learning and mobile phone data , 2017, EPJ Data Science.

[43]  Ram Dantu,et al.  Mobile Social Closeness and Communication Patterns , 2010, 2010 7th IEEE Consumer Communications and Networking Conference.

[44]  Scott A. Golder,et al.  Diurnal and Seasonal Mood Vary with Work, Sleep, and Daylength Across Diverse Cultures , 2011 .

[45]  Carolyn Penstein Rosé,et al.  Computational Sociolinguistics: A Survey , 2016, Computational Linguistics.

[46]  J. Wajcman,et al.  Families without Borders: Mobile Phones, Connectedness and Work-Home Divisions , 2008 .

[47]  Rohan Samarajiva,et al.  Who’s got the phone? Gender and the use of the telephone at the bottom of the pyramid , 2010, New Media Soc..