The Study Of Community Finding And Evolution Of Scientific Network

With the rapid development of science and technology,all kinds of complex networks come to our life,such as cooperation network,social network and academic reference network.People gradually found that the complex network has a number of important features,such as small world,scale-free,community structure etc.Finding the community structure of complex networks,and studying the evolution characteristics of complex networks have very important significance to study the function of the complex network,topology structure and properties,the hidden rule,and predict network behavior.Obviously,compared to other complex networks,scientific network has its own unique characteristics,not only lies in its node is scientific research workers,but also there are the complicated relationships between scientific research workers and the transmission and communication of knowledge in scientific network.Now,for the scientific research network group research mostly focused on software to realize,and not to specific research network analysis and the corresponding algorithm is proposed,for the study of the evolution characteristics of also is relatively broad,and there is no specific evolution characteristics is proposed.Combined the existing research results with the actual situation of science network,this study is being started for researching the algorithm of finding community structure and evolution characteristics.The purpose and significance is that understanding the internal structure of scientific network,mastering the evolution of scientific network,finding the community structure in the target field and analyzing the situation of research in the field and finding other valuable information.The main achievements of this paper include the following two aspects:First of all,the algorithm of finding community structure based on the similarity between nodes is built.This paper first analyzes the specific characteristics of scientific research network,mainly include the following aspects: First,nodes are scientific research workers,which have knowledge,cognitive,and emotional characteristics.Second,edge is formed by the relationship between the authors,such as cooperation and so on,and such relationship is not a simple connection and the connection,there is strength(that is,the weighted edge).Third,the scientific network is full of knowledge transfer and the exchange and the carrier is the literature,which describe the process and development of the society.Then,this paper built the algorithm of finding community structure on the basis of that.Secondly,the evolution model of scientific network is put forward.This paper mainly considered the evolution characteristics include the following: first,preferential attachment,describing the phenomenon that the new node is always closer to connect with the node whose degree is high.Second,aging effect,reflecting the contribution of the node and edge are diminishing along with the passage of time.Third,evolution of community structure,consider the evolution of community structure and study the evolution characteristics in the community structure.By the end of this paper,the main research results and conclusions,major contributions and limitations,as well as requiring work for further research are summarized.