Research On Key Node Recognition Method Based On Network Centrality Algorithm

Complex networks exist in various forms in life and production.Identifying the key nodes in the network can make a deeper understanding and control of the network.Centrality algorithm is mainly used in the identification of key nodes in the network.At present,the network centrality algorithm takes the single attribute of the nodes in the network as the research object to score the network nodes.In this paper,a key node identification method based on multi-attribute fusion is proposed by considering the global attribute and local attribute of network nodes.The key node identification method based on multi-attribute of fusion node proposed in this paper is suitable for undirected network,and the improved key node identification algorithm based on multi-attribute of fusion node is suitable for undirected weighted network,and is applied to supply chain network.In the unweighted network,based on the K＿Shell algorithm,the concept of KS deletion factor is proposed.The KS value and KS deletion factor are used as the global index input of the nodes in the network.At the same time,the degree of the node is used in the local index input of the node.The three indicators of the node are effectively integrated,and a method for identifying key nodes in the unweighted network based on the fusion of node multi-attributes is proposed,the key nodes in the network are identified,and the experimental results are verified by the SIR model.Experiments in four real networks prove that the algorithm in this paper is suitable for various networks and has high accuracy performance.In the weighted network,the method for identifying key nodes in the unweighted network based on node multi-attribute fusion is improved on the basis of adding the weight of the relationship between the nodes in the weighted network,and the method of using the KS value and KS deletion factor of the network node is The global input index,the degree and weight of the node and the index of the local input are effectively fused to achieve node multi-attribute fusion.A weighted network key node identification method based on node multi-attribute fusion is proposed to apply to the weighted network to realize the weighted network key node Recognition.Using the robustness of the network to evaluate the experimental results,the experiment proves that compared to the comparison algorithm,the improved key node identification method of fusion node multi-attribute is more accurate for the importance of nodes in the undirected weighted network.The weighted network key node identification method based on node multi-attribute fusion is applied to the supply chain network of a coal power plant,and the weight assignment method is used to assign weight to the supply chain network,identify important enterprises in the supply chain network,and provide The experimental results of the key node identification method of the chain network are compared with the experiment,and the robustness idea is used to verify the accuracy of the key node identification method based on the node multi-attribute fusion weighted network in the supply chain network.The experiment proves that the supply chain network of this article The key node recognition algorithm has high accuracy.This paper uses the identification of key network nodes as the guiding goal to improve the network centrality algorithm,and verify the accuracy of the algorithm in the unweighted network,and further add the weight of the weighted network to identify the key nodes in the weighted network,and finally integrate multiple nodes.The attribute-based key node identification method is applied to the supply chain network.Through real network experiments,the accuracy of the key node identification method based on node multi-attribute fusion proposed in this paper is proved.