Research On Node Importance In Complex Network Based On Information Fusion

Node importance ranking in complex networks is one of the research hotspots in the current academic field.The research of important node in complex networks is of great significance and wide application value for improving the reliability of complex networks.In recent years,many researchers have conducted in-depth analysis on complex networks and put forward many algorithms.This paper attempts to study the method of node importance ranking in multilayer complex network from the perspective of multi-attribute and multilayer complex network analysis.This paper proposes a node importance ranking algorithm based on degree and clustering coefficient.Firstly,the degree value and clustering coefficient of the nodes are calculated;Secondly,The entropy weight method is used to calculate the weight value corresponding to degree and clustering coefficient,and then calculate the weight factor of nodes;the nodes importance is calculated by combining weight factor,degree and clustering coefficient.the algorithm is tested with real complex network datasets.The results reflect that the algorithm proposed in this paper has certain advantages than some traditional ones.This paper proposes a node importance ranking algorithm in multilayer complex network in view of support vector machine.Firstly,the importance of nodes is evaluated by evidence theory;Secondly,this paper presents three simple indexes with low computational complexity,and uses support vector machine to find the mapping rules between the simple indexes and evidence theory evaluation;Finally,this paper uses the mapping model to calculate the importance of network nodes.This paper testes the algorithm with real complex network datasets.The experimental results of real network show the effectiveness of the algorithm: the evaluation based on evidence theory is in line with the reality.In this thesis,there are 19 figures,13 tables and 82 references.