Research On Network Analysis Method Based On The Features Of Core-periphery Matrix

In the process of continuous changes in social networks,from the perspective of microscopic changes,various information is transmitted and changed between individuals,so that the relationship between individuals will change differently according to the different information received.And this relationship may strengthen or break in the course of evolution.From the perspective of macroscopic changes,due to the nonlinear interaction of individuals in the network,the network will form a certain structure and function.Therefore,it is very necessary to analyze the network from two aspects.At present,researchers mostly focus on the analysis of a single network structure,or analyze the characteristics of a certain type of network.It is important to analyze these characteristics of the network,but the characteristics of the network are formed by the interaction of individuals.And the connection between individuals is a factor that directly affects the nature of individuals,so it is also necessary to study the coupling relationship between individuals.Moreover,the analysis of a single type of network cannot reflect the general phenomenon of social networks.In addition,the analysis of network structure in recent years is almost all screening individuals with similar characteristics from the network.In real society,each team usually includes individuals with different characteristics.Only by focusing on the diversity of the team can the tasks be successfully completed and the entire network will grow and develop.The core-periphery structure just satisfies this relationship.In addition,it is necessary to fully understand the characteristics of the network and the original hierarchical structure in the network in order to improve network security and strengthen network management.In 1966,J.R.Friedmann put forward the theory of core-periphery structure based on theories of regional development and economic growth.He believes that the country can be divided into a core area and a peripheral area.The core area is a city with rapid economic development and densely populated areas,and the peripheral area is a remote place with a relatively backward economy and sparse population.By 1999,Borgatti et al.proposed a matrix representation of the core-periphery structure.Later,many scholars have conducted a series of studies on the core-periphery structure of the network.The current research on this aspect is still stuck in dividing a single network into two parts(core and periphery).This article finds that it is not just a specific network or a certain type of network.This paper finds that it is not only a specific network or a certain type of network,but many social networks have a multi-layer core-periphery structure.In fact,not only many networks have a multi-layer core-periphery structure,but many relationships between individuals in the network also conform to the core-periphery structure.This paper analyzes the network from a new perspective,using the characteristics of the network matrix structure and combining some properties of the core-periphery structure.By calculating the intensity of the core-periphery structure relationship in the network,it is verified that there is a strong core-periphery structure relationship in the network.And this paper proposes for the first time a matrix-based network anatomical algorithm(MNAA)suitable for any network,which is used to dissect the layers of various networks and the core-periphery relationship between the layers,as well as the coreperiphery relationship between individuals in each layer.The content of this thesis is mainly divided in the following three parts:(1)Classify all networks.For any network,the connection between nodes is different.Because this research needs to be applicable to any network,we will classify the network at the beginning.Firstly,it is studied to store the connection relationship between nodes in a matrix,and sort the matrix of nodes to make the characteristics of the network matrix more obvious.By observing the 0,1 distribution in the matrix,it is found that the matrix type of the network can be summarized into three major types,so the network can be classified into three types according to the characteristics of the matrix.(2)Network layering.In the first step,this paper has divided the network into three types according to the characteristics of the network matrix.The second step is to layer the network in an appropriate way according to the characteristics of each type,so that the structure of the network is clearer and easier to analyze.After layering,through theoretical and experimental analysis,most network layers generally conform to the coreperiphery(CP)structure.(3)Analyze the relationship between nodes in the hierarchy.In social network analysis,this research first proposed and verified that the core-periphery relationship between nodes is also obvious,and such a relationship generally exists.Comparative experiments and verification experiments were carried out on a large number of generated networks and real networks.The experiment proves that there is a strong core-periphery(CP)structure relationship between the nodes in the layers and the nodes in the layers in the network.At the same time,through comparison,it is verified that the algorithm proposed in this paper has a great advantage.