Information Flow Simulations in Multi-dimensional and Dynamic Systems

Abstract The relevance of nodes with respect to the position they have in a network is often investigated centrality measures. In particular, cases where it specifically meaningful consider nodes’ ability cumulate and convey information, like economic systems, betweenness one most pertinent options because its underlying concept. However, this statistic presents two limitations. First, as relies on computation shortest paths, grounded binary topological evaluation: every time node not located path between other nodes, gains no score at all (even if just step longer). Second, does allow direct analysis multi-dimensional dynamic networks: has be computed dimension instant time, causes problems comparability case weighted connections. scope originality work design model that makes possible solve these issues. proposed Dynamic Multi-Layer Network (DMLN) allows structural representation properties interactions. Then, metric that, based Infomap random walks, assesses level information cumulated conveyed by any moment interaction. Importantly, performed without relying evaluation, jointly taking into account what occurred dimensions during entire period, which system observed. We present discuss an implementation ICT worldwide trade goods services period 2004–2014.