基于模体结构和度信息的关键节点组识别OA北大核心CSTPCD

In order to explore the impact of higher-order structures with smaller scales on key node group mining prob-lems and with the goal of optimizing network propagation,a key node group recognition algorithm was proposed based on motif structure and degree information.Firstly,the influence of nodes was evaluated based on the motif structure,and the core nodes of the motif structure were excavated.Then,the VIKOR method was used to fuse it with degree infor-mation.Finally,the seed exclusion algorithm was used to exclude the neighbors of the seed nodes,effectively reducing the problem of influence overlap.Based on the SIR propagation model,six different undirected networks were selected for comparison with four benchmark algorithms.The experimental results show that the proposed algorithm performs better in terms of accuracy and stability.