A novel algorithm for identifying key propagation nodes in complex networks based on Neighborhood Gravitational Structural Centrality

To comprehend the structure and operation of complex networks, it is essential to quantitatively evaluate the influence of nodes. In order to increase the precision of identifying important nodes in networks, we propose a unique influence identification algorithm called Neighborhood Gravitational Structural Centrality (NHGS). The algorithm measures the coupling of propagation between a node and its neighbors through Local Gravitational Strength (LGS), reflects the node’s global topological position using Structural Hybrid Score (SHS), and evaluates the node’s actual propagation range by introducing the Neighbor Influence Index (NII), which incorporates propagation overlap among higher-order neighbors. Finally, by integrating local connectivity, global structural position, and propagation overlap across adjacent layers, the algorithm provides a comprehensive evaluation of a node’s influence potential. Extensive experiments were carried out on eight real-world networks in order to verify the algorithm’s efficacy. Using Kendall’s tau, Jaccard similarity, and monotonicity metrics, comparisons were made with seven representative algorithms. The results of the experiment demonstrate that the proposed algorithm outperforms other algorithms in terms of differentiating ability and ranking accuracy, providing a quick and accurate way to assess node effect.