Influence maximization in social networks based on discrete harris hawks optimization algorithm

Influence Maximization (IM) is an important topic in the field of social network analysis, and is widely used in viral marketing, recommendation systems, rumor prevention and other fields. The meta-heuristic method has excellent performance because of high scalability and low complexity, however, the objective function of existing meta-heuristic methods can only be applied to propagation models with small probabilities. In addition, there is room for further improvement in performance of meta-heuristic methods. In order to solve the above problems, this paper transforms the influence maximization problem into an optimization problem and designs a novel objective function based on the six degrees of separation theory of social networks. Then, with the designed objective function, this paper discretizes the harris hawks optimization (HHO) algorithm by redefining the energy and location representation rules for influence maximization problem. Experimental results on eight real datasets demonstrate that the proposed objective function exhibits high accuracy and generality, suitable for various probability propagation models. With the exploration and exploitation process in steps, dynamically using different strategies various situations, the proposed DHHO algorithm exhibits better performance in dealing with influence maximization problem, outperforming the state-of-the-art methods.