2018/08/01 by Jorge Brea, Brea, Jorge, Javier Burroni +3
Physics and Astronomy · Social Sciences · #Complex Network Analysis Techniques #Computers and Society (cs.CY) #FOS: Computer and information sciences #Human Mobility and Location-Based Analysis #Machine Learning (cs.LG) #Opinion Dynamics and Social Influence #Social and Information Networks (cs.SI)
paper · pdf · doi:10.48550/arxiv.1808.00527
openalex publication_date 2018/08/01 · openalex created_date 2022/08/04 · openalex updated_date 2026/07/28
Over the past decade, mobile phones have become prevalent in all parts of the\nworld, across all demographic backgrounds. Mobile phones are used by men and\nwomen across a wide age range in both developed and developing countries.\nConsequently, they have become one of the most important mechanisms for social\ninteraction within a population, making them an increasingly important source\nof information to understand human demographics and human behaviour.\n In this work we combine two sources of information: communication logs from a\nmajor mobile operator in a Latin American country, and information on the\ndemographics of a subset of the users population. This allows us to perform an\nobservational study of mobile phone usage, differentiated by age groups\ncategories. This study is interesting in its own right, since it provides\nknowledge on the structure and demographics of the mobile phone market in the\nstudied country.\n We then tackle the problem of inferring the age group for all users in the\nnetwork. We present here an exclusively graph-based inference method relying\nsolely on the topological structure of the mobile network, together with a\ntopological analysis of the performance of the algorithm. The equations for our\nalgorithm can be described as a diffusion process with two added properties:\n(i) memory of its initial state, and (ii) the information is propagated as a\nprobability vector for each node attribute (instead of the value of the\nattribute itself). Our algorithm can successfully infer different age groups\nwithin the network population given known values for a subset of nodes (seed\nnodes). Most interestingly, we show that by carefully analysing the topological\nrelationships between correctly predicted nodes and the seed nodes, we can\ncharacterize particular subsets of nodes for which our inference method has\nsignificantly higher accuracy.\n