Identification of influential users with cost minimization via an improved moth flame optimization.

The issue of influence maximization has received great attention due to its application potential. Some traditional models used to solve the influence maximization problem only consider the maximum propagation range that the seed node set can reach but ignore the cost difference between the potential candidate nodes. This is not characteristic of real-world network behavior. Thus, in response to this research gap, a multi-objective optimization model based on maximizing influence spreading while minimizing cost is proposed in this paper. On the basis of maintaining the effective characteristics of non-dominated sorting moth flame optimization, the diversity weight and mutation mechanism are integrated into the algorithm to maintain the population diversity in the exploration stage. The evolutionary rules of the original moth flame optimization are redesigned to meet the desired needs of multi-objective influence optimization. By considering three types of real-world social networks, we show that our proposed method can generate a set of well-distributed Pareto optimal solutions.