BOSAM-based Topological Analysis And Modeling Of Scientific Collaboration Networks

Scientific collaboration networks (SCN) is one of social networks of complex networks. The study of its topology now is still on statistical level and people are always concentrating on describing the collaboration behavior of scientists. However, how do scientists collaborate with each other is still unknown.The primary achievements of this paper are as follows:1) The research on topology of SCN is analyzed, and the topology characteristics which can be observed from BOSAM bitmap is illustrated by skitter data and ER model.2) The topology characteristics of condensed matter collaborations are studied based on BOSAM, and two important characteristics of scientist collaboration behavior are described based on static BOSAM bitmap:I. New nodes do not tend to connect with strong nodes but to weak ones when joining the network.II. An interesting phenomenon called by us "combination of similar backgrounds" is discovered. It reflects that the collaboration behavior of scientists is concerned with power as well as backgrounds, that is nodes with stronger backgrounds tend to have stronger collaborators than with weaker backgrounds in details. It is a more clear explanation of scientist collaboration relationship.3) Based on the analysis of topology characteristics of SCN, we have constructed the topology model of SCN. The model in this paper is a dynamic growing and evolutionary model, which well concerns the node's power growing as the network grows, and three important characteristics of SCN evolution are 'described based on dynamic growth of the network scale:I. A new node connects with some node as well as the node's strongest neighbor.II. A new node prefers to connect with a weak node, and it does not have any opportunity to connect with a very strong node.III. The node which has already a strong collaborator tends to connect with the node which is stronger than it but has similar strength as it.The model well simulates the mechanism of SCN growing and scientist collaboration behavior, so it makes it clear on these aspects and testify some conclusion of us about SCN.