Data mining algorithms for modeling international scientific migration

The study of researchers’ migration is of scientific interest in terms of analysing an individual’s economic behaviour. The main challenge for these studies is the lack of reliable data on researchers’ migration. We collected the relevant data applying the methodology we developed that is implemented in software based on big data technology. This methodology relies on the analysis of information represented in abstract and citation databases of research literature: the data on scientific migration is obtained through examining changes in affiliation. The methodology has been tested on Scopus database: data on scientific migration of researchers employed by Ural Federal University were collected. The verification of the obtained data showed the high reliability of the results. Most researchers move to Western European countries and the United States (up to 72%). The main areas of emigrating researchers’ scientific interests are natural and technical sciences. Practical application of the brain sharing approach will minimize the negative impact of scientific migration on Russia’s scientific and technological security and will ensure realising the potential of the academic diaspora abroad for the development of Russian science and innovation. The practical value of the study lies in the possibility of applying its results for the development of the Russia’s science human resources potential.