A Novel Evolutionary Algorithm Based on Judgment-Rule Evolution Strategy for Structural Balance in Signed Social Networks

Structural balance in signed social networks has become a research hotspot for the task of finding the optimal network structure, which is able to make social systems reach a relatively stable and harmonious state. Usually, such a structural balance problem can be modeled as an optimization problem. The typical way of existing optimization methods for structural balance is to only minimize the total cost of changing the unbalanced edges in the whole network, but it neglects the load balance of changing edges between different nodes in signed social network, which may not fit the nature of the social systems. In this paper, we propose a novel structural balance model, which incorporates the total cost information of the changed edges in the whole network and the balance information of the changed edges of each single node. To optimize the novel structural balance model, we propose a Judgment-Rule-based Evolutionary Algorithm, called JREA, based on two significant essential features of the structural balance problem, i.e., the determination of node attribute depending on node’s degree and the attributes of its neighbour nodes. Extensive experiments are conducted on the four generated signed social networks and the three real signed social networks. Experimental results demonstrate the effectiveness and efficiency of the algorithm.