Stemming competitive influence spread in social networks through binary ions motion optimization

The rapid development of social networks has brought many conveniences, but it has also resulted in the wanton dissemination of negative information. Identifying key users in the network to block negative information in a timely and effective manner has become an urgent research task. For this purpose, this paper proposes a binary ions motion optimization algorithm to maximize the blocking of negative influence propagation under a competitive-based model. The algorithm adopts a degree-based heuristic initialization strategy by recoding search agents and blocking diffusion channels based on the negative seed location. To overcome the lack of crystal phase search ability, a crossover mechanism of anions and cations is introduced, which accelerates convergence and facilitates the discovery of optimal solution. Finally, the effectiveness of the proposed algorithm is demonstrated on real networks and synthetic networks, showing significant advancements compared to other algorithms.