Social Network Analysis in Scientometrics

In this paper we report on the results of our current attempt to enhance scientometric measurements by employing social network analysis and mining methods. We begin by recalling of our previous work on the collection of a rich data on the social network of scientific collaboration. Then, we proceed to the description of the enhancements to the dataset. Most importantly, we report on the three separate results obtained from the extended social network. The analysis of the triad closure consensus in the dataset reveals interesting patterns regarding the underlying nature of scientific collaboration. Even more evident are the results of the betweenness centrality analysis, where a periodic pattern emerges both in co-authorship and co-participant networks. Finally, we conclude with the introduction of a complex model of scientific career development which uses conditional probability sequential patterns.