Intimacy Analysis And Visual Display Of Scientific Research Cooperative Community

In recent years,with the increasing complexity of scientific research,scientific research work has gradually turned to the way of cooperation.Scientific research cooperation helps to promote knowledge sharing,resource sharing and improve the efficiency of scientific research output.Therefore,it is of great significance to study the internal organizational structure and evolution mechanism of scientific research cooperation,which plays a vital role in the management of scientific and technological work and the formulation of scientific and technological policies.This paper relies on the "Hebei Science and Technology Innovation Big Data Public Service Platform" and takes the scientific research papers in the output of scientific and technological achievements as the research object.Through the construction of scientific research cooperation network,this paper focuses on three aspects: community division of scientific research cooperation network,visualization layout algorithm and evaluation method of core nodes.The main research work of this paper is as follows.(1)A core node evaluation model based on network topology and heterogeneity information is constructed.From the point of view of network topology,the evaluation method of core nodes is given by considering the size of node neighbors and the degree of tightness between node and neighbor nodes.The heterogeneity information of nodes mainly reflects the influence of the number and quality of papers published by scientific research institutions on the importance of nodes.From the perspective of the impact of papers,the evaluation method of core nodes is given.(2)The Louvain community partition algorithm is improved.Considering the tightness of cooperation between scientific research cooperation institutions,the measurement methods of node intimacy are defined from the point of view of node similarity based on network topology structure and node similarity based on domain.On this basis,the network of scientific cooperation is weighted by the value of intimacy.In the second stage of Louvain algorithm,the core nodes of scientific research cooperation network are regarded as the central nodes of clustering.Experiments show that the algorithm can divide the scientific research cooperation network community from the characteristics of node domain and network topology structure,and can quickly find the scientific research team in the network.The result of the division is reasonable.(3)This paper presents a layout algorithm for displaying the community structure of scientific research cooperation network.On the basis of FR algorithm,the algorithm optimizes the gravitational model by using the intimacy value between nodes,and promotes the closest Euclidean distance between nodes with larger intimacy value in the visual display process.At the same time,community repulsion should be added among research communities to prevent excessive closeness between communities.The experimental results show that compared with FR algorithm,this method can show the internal structure of scientific research community more clearly.