Dresden 2017 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 21: Social Systems II
SOE 21.2: Talk
Thursday, March 23, 2017, 17:15–17:30, GÖR 226
Dynamic Content-Communities on Social Networks — •Philipp Lorenz1, Frederik Wolf1, Jonas Braun2, Philipp Hövel1, Colin Bauer3, Julien Siebert3, and Vitaly Belik4 — 1TU Berlin, Hardenbergstraße 36, 10623 Berlin — 2HU Berlin — 3Zalando SE — 4FU Berlin
One of the best studied property of real world networks is their community structure, which represents their composition of dense subnetworks. They can be overlapping [1], hierarchical [2] and temporal [3]. In human social networks all of these properties come into play [4] and can be found on different levels.
On Internet platforms content is posted and within that, trends can be captured as communities of linked topics. They behave highly dynamical, they are born, merge, split or grow and shrink, so capturing them required new methods for temporal community analysis.
Such clusters of topics can move and spread rapidly on social networks. Since these movements depend strongly on the underlying topology, we created multilayer networks, which link contents to users and uncover their relations and interplay.
[1] Palla, G. et al., Nature 435, 7043 (2005) [2] Peixoto, T. P., Phys. Rev.X 4, 011047 (2014) [3] Palla, G. et al., Nature 446, 664 (2007) [4] Sekara, V. et al., PNAS 113, 36, (2016)