Research On The Sequence Algorithm Of The Nodes Importance In Complex Network

In our daily life,Internet,traffic network,social network and other network systems are ubiquitous.By analyzing these basic networks,we find that they not only bring convenience to human beings,but also pose potential safety problems.For example,traffic jams,Internet crashes,power grid failures and more.Therefore,it is necessary to discover important nodes in the network.At present,some achievements have been made in the evaluation of the nodes importance in complex network.However,due to the large number of nodes in a complex network and the complex structure and function,the existing methods have some defects.Based on the existing methods,this paper presents two new algorithms for ranking the importance of nodes,which are the node importance measurement method of complex networks based on multi-index and node deletion integration and the node importance sorting algorithm of improved node contraction method in weighted networks.The node importance measurement method of complex networks based on multi-index and node deletion integration is applicable to an undirected and non-weighted networkproposed.The algorithm gives the importance index of nodes through the entropy weight TOPSIS method which includes four properties: degree centricity,betweeness centrality,closeness centrality and clustering coefficient.The PR value of each node is given by Page Rank algorithm,and then the parameters ? and? to calculate each node's comprehensive importance rankings.Then,the importance of nodes is dynamically measured through the improved node deletion method.This algorithm overcomes the one-sidedness of the single-metric measurement and the subjectivity of the setting of attribute weights.It comprehensively considers the impact of network structure changes and important neighboring nodes on the ranking results.Simulation results on the ARPA network show that the proposed method can effectively reduce the reliability of ranking of node importance due to the change ofnetwork structure and the influence of adjacent nodes.The node importance sorting algorithm of improved node contraction method in weighted networks applies to undirected weighted network.This algorithm objectively assigns weights to the transformed network edge based on the multi-attribute entropy-weighted method,and calculates the aggregated value of the transformed network as a part of the importance measure of the original network node.Then,the proportional coefficients ? and ? are introduced to linearly calculate the importance of the node itself and the importance of the edges.Finally,the weighted network's node importance ranking results are obtained.The algorithm avoids the subjective error caused by human empowerment and the influence of the edge weight difference.Through case analysis and comparative experiments with existing methods,it shows that the method is intuitive and effective,and provides a reference for the importance of nodes in complex weighted networks.