A Swarm Intelligence Approach for Effective Influence Maximization in Online Networks

Numerous studies addressing the traditional influence maximization problem have prioritized maximizing influence spread. This research presents MOCIA, a unique discrete evolutionary framework-based multi-objective crow intelligence system that aims to attain least cost. Optimizing MOCIA's parameters using a dynamic control methodology and a random walk method derived from black hole dynamics enhances the convergence efficiency. Using actual social networks, we compared our algorithm to MOCSA. The proposed MOCIA frequently outperforms benchmark algorithms, with a significant reduction in processing time and an increase in objective function value by more than 20%. Furthermore, MOCIA effectively balances exploration and exploitation during the optimization process. It exhibits consistent performance across diverse network topologies, highlighting its adaptability. The algorithm’s modular design also facilitates integration with other influence-aware applications, offering broad utility in real-world scenarios.