Research And Application Of Complex Network Projection Model

Complex network is a type of network composed of a large number of individuals and close connections and interactions between individuals.In recent years,the development speed of complex network theory has been amazing,attracting the attention of many researchers.Researchers applied it to the common social networks,biological networks,and transportation networks in our daily lives.In order to better understand the actual network(system)in life,it is necessary for us to strengthen the research on the theory and application of complex networks.This will be very helpful for us to have a deeper understanding of the actual network in life.When performing network analysis on real systems in various fields,the establishment of network models can make our work more efficient.Many practical systems in our lives can be characterized by bipartite networks,such as gender networks for men and women,network for athletes' participation,network for college students' course selection,and so on.Based on the work done by predecessors,this paper optimizes the original bipartite network projection model.In the bipartite network =((3,(4,),this model can help us obtain the importance of the nodes of the set (4 in the perspective of the set (3.Specifically,when the nodes of set (4 have the same initial resources,based on their own number of edges and the weight of the edges,the initial resources are flow from the nodes of set (4 to the corresponding nodes of set(3.At the next moment,the resources are again flow the nodes of set (3 back to the nodes of set (4.After multiple projection processes,the final resources of the nodes in the set (4converge to a constant value.The result of this model depends on the topology of the network,the weight of the connected edges,and the initial resource value of the set (4nodes.That is,the more the number of edges connecting the node,and the greater the weight of the associated edge,the higher the final resource of the node,and the more important the node is from the perspective of the set (3.The value is called the importance of the node in this paper.Since the financial field has been a hotspot in complex network research in recent years,this paper chose to apply this model to the actual financial system.Specifically,this paper propose a fund-stock network projection model to study the investment rules of funds.This model can calculate the importance of each stock,which can effectively quantify the relative investment intensity of the fund in stocks.Starting from the importance of stocks,we studied the investment level,investment distribution and investment tendency of funds;this paper proposed an investment strategy based on the importance of stocks,which provided a certain reference for individual retail investors.Then,in order to study the group holding behavior of funds,based on the fund-stock network projection model,this paper proposed a weighted projection method based on the similarity between stock nodes,and established a weighted stock network.this paper carried out community detection on the network and gave a distribution map of its communities,found the key communities in the network,and further analyzed the evolution of the importance of key communities and the evolution of the importance ratio of industries in key communities.