Dynamics of regional multilinks in research innovation temporal networks

¹ Department of Physics, University of Thessaloniki - Thessaloniki, Greece
² Center of Complex Systems, University of Thessaloniki - Thessaloniki, Greece
³ Department of Physics, International Hellenic University - Kavala, Greece

Received: 20 February 2020
Accepted: 27 April 2020

Abstract

In this paper, we examine the evolution of a temporal multiplex innovation network, and develop a kinetic model that describes the dynamical process behind its growth. The multiplex consists of two collaboration network layers. The nodes of the first layer are the European regions of the participants of the EU Framework Programme (FP) projects, and the nodes of the other layer are the European regions of the patent inventors. A link between two regions exists, when scientists associated with those regions have collaborated in an FP project or inventors in a patent. The analysis has been conducted using the notion of multilinks, which essentially describes differences and similarities in the connectivity between identical nodes in both layers. A sliding windows method was employed in order to study the network in various time periods, and all multilinks were calculated in each window. All three types of multilinks were studied (Framework Programme (1, 0), patents (0, 1) and common ones (1, 1), where links between the same regions exist in both layers). Results indicate that all multilink types exhibit a roughly similar growth pattern through the course of 16 years with few observable changes. The results also point out that patents are the driving force for the creation of common multilinks early on in time, while this is reversed later on. We suggest a simple kinetic model of 3 differential equations and 6 parameters that adequately describes the system dynamics. The parameter values exhibit a surprisingly small variation for large periods of time. We believe that this model could easily be extended to other systems with links which are added or removed (birth/death), or even to multiplex networks with more layers.