According to the Concept of the State Program of the Russian Federation “Education Development” until 2030, in order to create advanced professional training of specialists, it is necessary to more fully use the mechanisms of social partnership, which allows to coordinate the diverse needs of the labor market and the academic community. However, for a more reasoned inclusion of social partners in the process of professional training, a multidimensional model specifying the goals and functions of all partnership participants at each stage of students’ education in higher education is needed.
The aim of the work is to substantiate the taxonomic model of professional training of future teachers in social partnership of pedagogical university and preschool educational organization.
The methodology and methods of the research are based on the system, competence and qualimetric approaches, which allow to determine the agreed with social partners competence-oriented goals of professional training of future teachers in the system of social partnership of pedagogical university and preschool educational organization using the method of group expert evaluations. The structure and content of the taxonomic model of students’ training in social partnership between a pedagogical university and a preschool educational organization are determined by the method of group expert evaluations. The proposed taxonomic model can be useful for the university administration and academic staff in designing the educational process within the framework of individual educational trajectories of students, taking into account the needs of the regional labor market and the resources of social partners.
Keyword(s) : pedagogical university
Predicting Student Employment in Teacher Education Using Machine Learning Algorithms
One of the solutions to the problem, when not the best graduates enter the pedagogical profiles and after graduation are employed in the education system, is the prediction of professional orientation even at the stage of the student choosing their further professional trajectory. To solve this problem, the purpose of the study is to develop and experimentally prove the effectiveness of using a program for predicting the employment of students of a pedagogical university based on the introduction of various machine learning algorithms. Using a random selection of students, the collection and processing of their questionnaires (n=205) in 2011-2016 were carried out. Various machine learning algorithms were used to create the program: decision trees, logistic regression, and catboost. In the course of the experiment, the data of the questionnaires were loaded into the program for its training according to various algorithms, in order to ultimately obtain a finished intellectual product with the ability to predict the employment of graduates. In the final comparison, the program developed on the “decision trees” algorithm made only 2 out 19 questionnaires and 7 out 61, which was the best result – 89%. The implementation of this algorithm makes it possible to most accurately, with the least percentage of errors, identify students who will not be employed in the future according to their profile of study or not employed at all. Thus, the study developed an intelligent program that allows one to instantly process data and get an accurate forecast of employment with only a small probability of error.