Quantcast

Dynamic topologies of activity-driven temporal networks with memory

Research paper by Hyewon Kim, Meesoon Ha, Hawoong Jeong

Indexed on: 21 Nov '17Published on: 21 Nov '17Published in: arXiv - Physics - Statistical Mechanics



Abstract

We investigate the interplay of the time resolution and memory in dynamic scaling properties of temporal networks, where we present an activity-driven network model with memory. We also show how to control the time resolution and memory. Temporal networks are time series of static subnetworks. The time resolution in analyzing networks plays an important role in determining the dynamic topologies of networks. Performing the random-walk (RW) process on coarse-grained networks, we explore the role of memory in the diffusion properties on the RW process. In particular, we focus on the evolution of dynamic topologies in temporal networks with and without memory. Based on the temporal percolation concept in time-aggregated networks, we discuss the relation between the growth patterns of the largest cluster size and the diffusion properties of the RW process. Finally, we propose the finite-size scaling (FSS) form in the dynamic topologies of temporal networks, which are numerically confirmed with analytic conjectures.