From temporal network data to the dynamics of social relationships

摘要: Networks are well-established representations of social systems, and temporal networks widely used to study their dynamics. Temporal network data often consist in a succession static over consecutive time windows whose length, however, is arbitrary, not necessarily corresponding any intrinsic timescale the system. Moreover, resulting view evolution unsatisfactory: short contain little information, whereas aggregating large blurs Going from meaningful evolving representation therefore remains challenge. Here we introduce framework that purpose: transforming into an weighted where weights links between individuals updated at every interaction. Most importantly, this transformation takes account interdependence relationships due finite attention capacities individuals: each interaction two only reinforces mutual relationship but also weakens with others. We concrete example such apply it several sets interactions. Using contact collected schools, show how our highlights specificities structure organization. then synthetic perturbation set interactions group baboons possible detect on wide range timescales parameters. Our brings new perspectives analysis networks.