Measuring Diversity of Research by Extracting Latent Themes from Bipartite Networks of Papers and References

摘要

Inspired by the hypothesis that the diversity of research might decline as a result of new science policy measures we explore the potential of bibliometric measures for analysing the diversity of research at meso- and macro-levels of (national sub-) fields, countries, and organisations. Our aim is to render changes in the diversity of research landscapes measurable and therefore comparable in time series as well as between different countries. We discuss different methodological approaches and some results based on amethod that extracts latent themes from bipartite networks of research papers and their cited references by singular value decomposition of the citation matrix.