A hierarchical structure for opinion convergence in multi-agent networks

This study extends Taylor's model of opinion dynamics by introducing a hierarchical framework that refines the characterization of opinion convergence and containment in multi-agent systems. The proposed model structures agents into multiple hierarchical levels, where the convergence region of each level is influenced by the opinions of agents in the upper level. This organization provides a more detailed understanding of how opinions evolve in networks influenced by stubborn agents. Furthermore, the model is extended to incorporate time-varying stubborn opinions, enabling the analysis of dynamic external influences and their impact on opinion formation. This enhancement makes the framework more applicable to real-world scenarios, where leadership positions or external biases evolve over time. The effectiveness of the proposed model is validated through numerical simulations.