Predicting labor market trends through the analysis of open data from job applicants in social networks

The article examines trends in the supply of labor market, including the reduction of supply, "aging" workforce, and widening gap between professional and qualifications of job seekers and vacancies.Economic sociology methodology allowed to identify and analyze the influence of quality variables on current state and development trends of supply.Social networks are proved to be a valuable source of information about professional choice.Special methods for obtaining information from applicant's route maps and processing with machine learning models allow to build analytical tools for identifying social and behavioral characteristics, predicting future choices.Quality metrics for machine learning models are presented and compared.Results indicate high quality of forecasting tools, which can be used to adjust supply.Practical significance lies in improving career guidance, prediction of applicants' choice, optimization of the allocation of university places to students, assessment of admission stress index.Our analytical tools allow to assess how the planned indicators correspond with demand for education and to analyze the effectiveness of the mechanism for informing students about market trends.