Globalization of the scientific community has led to an unprecedented mobility of highly qualified personnel, creating complex challenges for national innovation systems. The study focuses on the contradiction between the positive aspects of academic mobility and its destructive impact on donor countries in the context of the “brain drain”. A client-server software package for inductive modeling of scientific personnel migration processes based on big data has been developed. The method of group accounting of arguments with the construction of hierarchical polynomial models selected according to the regularity criterion has been used. Migration factors have been formalized through the functions of attraction and repulsion. Structural heterogeneity of migration flows with the dominance of STEM specialists has been revealed. A hierarchy of migration determinants has been established: career and infrastructural factors are more significant than economic ones. A nonlinear model of migration probability has been obtained. Validation by the k-fold cross-validation method confirmed the stability of the model. The effectiveness of inductive modeling for analyzing migration processes was proven. A toolkit for quantitative assessment of human capital management policies was proposed.