Inferring friendship network structure by using mobile phone data Export

@article{citeulike:5455141, abstract = {10.1073/pnas.0900282106 Data collected from mobile phones have the potential to provide insight into the relational dynamics of individuals. This paper compares observational data from mobile phones with standard self-report survey data. We find that the information from these two data sources is overlapping but distinct. For example, self-reports of physical proximity deviate from mobile phone records depending on the recency and salience of the interactions. We also demonstrate that it is possible to accurately infer 95\% of friendships based on the observational data alone, where friend dyads demonstrate distinctive temporal and spatial patterns in their physical proximity and calling patterns. These behavioral patterns, in turn, allow the prediction of individual-level outcomes such as job satisfaction.}, author = {Eagle, Nathan and Pentland, Alex S. and Lazer, David}, citeulike-article-id = {5455141}, citeulike-linkout-0 = {http://dx.doi.org/10.1073/pnas.0900282106}, citeulike-linkout-1 = {http://www.pnas.org/content/early/2009/08/14/0900282106.abstract}, citeulike-linkout-2 = {http://www.pnas.org/content/early/2009/08/14/0900282106.full.pdf}, citeulike-linkout-3 = {http://view.ncbi.nlm.nih.gov/pubmed/19706491}, citeulike-linkout-4 = {http://www.hubmed.org/display.cgi?uids=19706491}, comment = {Great paper for me to reread for inspiration, since their data \& task are a bit like mine. Given: 2 groups of people with cell phones. Record (every 5 minutes for a school year): closest cell phone tower, and whether you're in proximity to anyone else in the study. Analyze: -whether proximity patterns correlate with self-reported proximity (--> yes, with memory for about 1 week, and better memory for friends) -whether proximity patterns predict friendship (--> yes, and in fact they seem to predict non-reported friendships). via simple features, e.g.: amt of time spent together during work hours (neg correlation), outside of work hours (pos correlation); direct phone calls (pos correlation). -whether friendship patterns predict job satisfaction (--> yes, having friends at work makes you happy; calling your friends from work indicates you're not) Creepy: "There is no technical reason why data like these cannot be collected from millions of people throughout the course of their lives." but interesting save: ". . . the potential for achieving important societal goals, from urban planning to public health, is considerable." Similar goal to me in: "Leveraging these behavioral signatures to accurately characterize relationships in the absence of survey data also has the potential to enable the quantification \& prediction of macro social network structures . . ." Yes, data available, and I've downloaded it.}, day = {8}, doi = {10.1073/pnas.0900282106}, journal = {Proceedings of the National Academy of Sciences}, keywords = {social\_network\_analysis, tribes}, month = {September}, number = {36}, pages = {15274--15278}, posted-at = {2009-09-22 22:35:11}, priority = {2}, title = {Inferring friendship network structure by using mobile phone data}, url = {http://dx.doi.org/10.1073/pnas.0900282106}, volume = {106}, year = {2009} }

TY - JOUR ID - citeulike:5455141 L3 - citeulike-article-id:5455141 N2 - 10.1073/pnas.0900282106 Data collected from mobile phones have the potential to provide insight into the relational dynamics of individuals. This paper compares observational data from mobile phones with standard self-report survey data. We find that the information from these two data sources is overlapping but distinct. For example, self-reports of physical proximity deviate from mobile phone records depending on the recency and salience of the interactions. We also demonstrate that it is possible to accurately infer 95% of friendships based on the observational data alone, where friend dyads demonstrate distinctive temporal and spatial patterns in their physical proximity and calling patterns. These behavioral patterns, in turn, allow the prediction of individual-level outcomes such as job satisfaction. IS - 36 JF - Proceedings of the National Academy of Sciences EP - 15278 TI - Inferring friendship network structure by using mobile phone data VL - 106 SP - 15274 KW - social_network_analysis KW - tribes N1 - Great paper for me to reread for inspiration, since their data & task are a bit like mine. Given: 2 groups of people with cell phones. Record (every 5 minutes for a school year): closest cell phone tower, and whether you're in proximity to anyone else in the study. Analyze: -whether proximity patterns correlate with self-reported proximity (--> yes, with memory for about 1 week, and better memory for friends) -whether proximity patterns predict friendship (--> yes, and in fact they seem to predict non-reported friendships). via simple features, e.g.: amt of time spent together during work hours (neg correlation), outside of work hours (pos correlation); direct phone calls (pos correlation). -whether friendship patterns predict job satisfaction (--> yes, having friends at work makes you happy; calling your friends from work indicates you're not) Creepy: "There is no technical reason why data like these cannot be collected from millions of people throughout the course of their lives." but interesting save: ". . . the potential for achieving important societal goals, from urban planning to public health, is considerable." Similar goal to me in: "Leveraging these behavioral signatures to accurately characterize relationships in the absence of survey data also has the potential to enable the quantification & prediction of macro social network structures . . ." Yes, data available, and I've downloaded it. AU - Eagle, Nathan AU - Pentland, Alex AU - Lazer, David PY - 2009/09/08/ UR - http://dx.doi.org/10.1073/pnas.0900282106 ER -