The Research Of Measurement Model About Complex Networks Node Importance

A large number of studies on actual network have showed that complex networks have different heterogeneous topology structures essentially,which determines that the position of each node in the network is not equal.For example,from the Communication Network to the Traffic Network,from the Protein Network to the Information Network,each individual's importance is distinct in the network.For the reason that identifying the important nodes in the complex networks plays an important role in selecting the attack strategy and optimizing the allocation of resources,how to use the method of quantitative analysis to identify the important nodes in the complex network is one of the urgent problems that need to be solved.Nowadays,there are two outstanding problems in the evaluation of node importance in complex networks.The one is that it is unilateral and limited to use the single index to evaluate the node importance in different types of complex networks.The other one is that the researches abut the node importance measurement in the complex networks mainly focus on the unweighted-undirected network.In order to solve these two problems mentioned above,this paper puts forward the node importance comprehensive evaluation model and the node importance measurement model in four types of complex networks.The main contents of this paper are showed as follows:(1)Considering that the complex networks' topology structure is not the same,and the index of evaluating the node importance has its limitations,it is more advisable to evaluate the importance of nodes in the real complex networks from different angles,using the node's multiple properties.Based on this idea,a comprehensive evaluation model which is used to evaluate the importance of nodes is proposed in this paper.Comprehensive evaluation model combines the advantages of multiple indices,which can identify the important nodes in different types of complex networks.Finally,the result of complex networks attack shows that the comprehensive evaluation model is more comprehensive and stable compared to using a single index.(2)According to the intensity and direction of the edge,complex networks can be divided into the unweighted-undirected network,the weighted-undirected network,the unweighted-directed network and the weighted-directed network.However,most of the index and algorithm can only be applied in dealing with the unweighted-undirected network.In view of the four different types of complex networks,this paper puts forward four kinds of suitable node importance measurement model respectively.The main idea in these four measurements is that the node importance is decided by their own properties and all the other node's contributions together.Although the first-order neighbor nodes have the most significant influence to the evaluated node,the m-order nodes are also linked with this node,and the influence of the m-order neighbor nodes cannot be ignored.Finally,in a large number of simulation experiments,the effectiveness and accuracy of the four models are verified by comparing with the classical indices in the "accepted" complex networks.In a word,aiming at these two problems existing in the current complex networks node importance evaluation,this paper puts forward a corresponding new evaluation method.The study has showed that the new evaluation method can identify the important nodes in the complex networks better.