A growing complex network design method with an adaptive multi-objective genetic algorithm and an inner link restructuring method

Many systems that can be modeled using network structures appear in various fields such as informatics, social science, economics, ecology, biology, and engineering. If these systems can be modeled as complex network systems, a complex network design method that finds a desired network structure can become one of strong tools in large-scale system designs. Conventional complex network design methods generally tackle static networks. On the other hand, a growing complex network design method has been proposed. In this method, the network is grown with determining the link structure of the newly-added node in order to obtain a desired network. In this design method, it is expected that characteristics of the optimization problem solved in the algorithm vary along with growth of the network. However, parameters of the optimization method are fixed; hence, its search process may be inefficient. In the design method, connection targets of a new node are only determined, and the structure of the existing network is not changed. Therefore, if the initial network has a large number of nodes, the desired network may not be obtained by using the design method. In this study, we propose to introduce an inner link restructuring method and an adaptive MOGA, that is, an adaptive SPEA2 into the growing complex network design method to deal with foregoing issues. Then, we confirm the effectiveness of the proposed method through numerical experiments.