Abstract
Abstract
Influence maximization (IM) aims to select a set of influential nodes in social networks to maximize information diffusion. However, existing methods struggle to balance solution accuracy and computational efficiency, restricting their practical deployment as network size increases. This paper proposes a Discrete Dung Beetle Optimizer for the IM problem. We discretize the Dung Beetle Optimizer and introduce an SD-based initialization to enhance population diversity and convergence speed. During search, the population is divided into functional roles with tailored discrete update rules, and crossover–mutation operators are incorporated to strengthen global exploration. An adaptive local search mechanism is further designed to refine solutions by combining node degree and iterative progress. Experiments on six real-world social networks show that DDBO achieves stronger influence spread and higher time efficiency than state-of-the-art methods, effectively balancing accuracy and cost.