Motif dynamics in signed directional complex networks

Abstract

Complex networks evolve and vary their structure as time goes by. In particular, the links in those networks have both a sign and a directionality. To understand their structural principles, we measure the network motifs, which are patterns that appear much more than one would expect in randomized networks, considering both link properties. We propose motif dynamics, which is a study to investigate the change in the number of motifs, and applied the motif dynamics to an open evolving network model and empirical data. We confirm that a non-cyclic motif has a greater correlation with the system size than a cyclic structural motif. Furthermore, the motif dynamics can give us insight into friendship between freshmen in a university.