Analysis And Research On The Collaboration Network Of Scientific Project

With the continuous development of modern science,scientific collaboration has become the mainstream of scientific research.There are two forms of scientific collaboration,one is the co-authorship,the other is scientific project collaboration.Based on cooperation relation between researchers,this paper is mainly to study the characteristics of persons and subjects in the scientific project collaboration network and to locate the scientific research teams hidden in the network.Firstly,this paper builds a collaboration network of scientific projects based on the scientific project dataset.The nodes in the network represent the researchers and the edges represent cooperative relations among the researchers.Secondly,Using the centrality analysis,network density analysis,clustering coefficient analysis,small world characteristic analysis and scale-free characteristic analysis in the social network analysis method to make a deep research on the collaboration network of scientific projects.Finally,the scientific research teams in the collaboration network of scientific projects are excavated by using the clustering subgroup analysis algorithm and the community mining algorithm in the social network analysis method.The main research work and innovation of this paper has the following three parts:1)Most of the researches on scientific collaboration are the co-authorship,and there is no report on the researches of scientific project collaboration.In this paper,scientific project collaboration is used to analyze the cooperative relations among the researchers.And provide a new idea for the scientific collaboration.2)This paper analysis the collaboration network of scientific projects by using the central analysis,network density analysis,clustering coefficient analysis,small-world characteristic analysis and scale-free characteristic analysis in the social network analysis method,and analysis the scientific project cooperation among the researchers,and analysis the proportion of the core researchers and their age composition,and analysis the small-world characteristics and scale-free characteristics of the collaboration network of scientific projects.3)Combining the cohesive subgroup analysis algorithm with the merging strategy,a coalescing clustering algorithm is proposed to mine the scientific research team hidden in the collaboration network of scientific projects,and this paper compares the proposed algorithm with the FN Algorithm and LPA algorithm,the results show that the scientific research team based on the coalescing clustering algorithm proposed in this paper is more suitable to the actual situation.