2017/01/20 by Paolo Barucca, Fabrizio Lillo, Barucca, Paolo +5
Physics and Astronomy · Psychology · #Complex Network Analysis Techniques #FOS: Computer and information sciences #FOS: Physical sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Mental Health Research Topics #Opinion Dynamics and Social Influence #Physics and Society (physics.soc-ph) #Social and Information Networks (cs.SI)
paper · pdf · doi:10.48550/arxiv.1701.05804
openalex publication_date 2017/01/20 · openalex created_date 2022/09/29 · openalex updated_date 2026/07/28
We study the inference of a model of dynamic networks in which both\ncommunities and links keep memory of previous network states. By considering\nmaximum likelihood inference from single snapshot observations of the network,\nwe show that link persistence makes the inference of communities harder,\ndecreasing the detectability threshold, while community persistence tends to\nmake it easier. We analytically show that communities inferred from single\nnetwork snapshot can share a maximum overlap with the underlying communities of\na specific previous instant in time. This leads to time-lagged inference: the\nidentification of past communities rather than present ones. Finally we compute\nthe time lag and propose a corrected algorithm, the Lagged Snapshot Dynamic\n(LSD) algorithm, for community detection in dynamic networks. We analytically\nand numerically characterize the detectability transitions of such algorithm as\na function of the memory parameters of the model and we make a comparison with\na full dynamic inference.\n