The Effect Of Data Depth On Community Structure Detection In Complex Networks

In real life many complex systems can be abstracted into complex networks,and the use of complex network theory can be a good help to understand the system characteristics.Community structure as one of the most important features of complex networks,which reveals the relationship between structure and function in the network,reflects the deeper characteristics of the network.The current real-life network is getting bigger and bigger with the exchange of information network size,so how to find and use the network association structure accurately has become a hot research problem of complex systems at present.The community in the network is the result of individual selection.This paper starts from the reasons for the formation of the real network community and defines the similarity of nodes so that nodes can form a node pair by selecting the most similar nodes to form a community.After the nodes form a node pair,the order of the node pair is changed,and the results of the Monte Carlo model are used to find that the node will preferentially select the node with a higher degree to form a node pair.Based on this result,a pairing algorithm based on the maximum node degree and similarity is proposed.The algorithm assumes that the formation of communities is the result of nodes selecting the most similar nodes,and that pairs of nodes with greater similarity and degree will preferentially form communities,and each node can have the right to choose multiple communities,thus discovering the overlap of communities in the network structure,which reasonably explains the physical meaning of overlapping nodes,and the algorithm proposed in this paper can well discover the overlapping community structure of the network.Since the overlapping nodes have the same degree of connection with the community it belongs to,in order to find more deep overlapping nodes,increase the similarity series and continue to optimize the community by using the multi-level most similarity.Find out the deep overlapping structure and the sub-community structure of overlapping communities in the network,and explain the evolution process of the network from node pairs to small communities and then to overlapping communities.The model in this paper reveals the deep overlapping structure of the network and discovers the substructure of the community,which has a good interpretation and analysis effect on the actual network structure.Considering that the different selection and pairing methods of nodes will have an impact on the community structure of the network,the nodes are respectively restricted to select only one node and two nodes to redivide the community.This paper first restricts nodes to select only one most similar node.Through Monte Carlo simulation,it is found that this pairing method discovers the small community structure of the network,and most of these small communities are at the edge of the network.At the same time,it explains that the network community structure will produce diversity due to the selection of nodes,which is also one of the important characteristics of real networks.Then the restricted node can choose two similar nodes to form a node pair.Using Monte Carlo,we found a stable community structure of the network that only needs part of the data.When the node takes the degree of the node as the probability,it will also make the node closer to the center of the network.This paper explains the reasons for network diversity by analyzing two different pairing selection methods,and at the same time obtains the influence of node pairing methods on network community structure.