Abstract:
This study aims to identify discrepancies between the requirements of employers and the labor market, educational standards and preferences of students in the field of information technology using large language models. Based on the texts of vacancies, using the open APIs of the Head Hunter service, an algorithm has been developed for compiling an average vacancy on the market, reflecting the most frequent requests from employers. To verify compliance and identify discrepancies between the requirements of the labor market and the actual knowledge of students, this study proposes an algorithm for forming a target vacancy according to the work program of disciplines and the expected vacancy by students based on a survey of knowledge, skills, current and planned to study, skills of students. The paraphrase-MiniLM-L6-v2 language model was used to compare vacancies compiled in the framework of this study. Results: The analysis revealed significant discrepancies between market requirements for IT specialists and educational programs in the field of information technology, as well as a discrepancy between students' expectations and competencies demanded by employers (compliance 65.3%). The results emphasize the need to adapt educational programs in the field of information technology to rapid changes in the labor market and demonstrate the effectiveness of using large language models in automating this process.
Keywords:Information technology, large language models, higher education, competencies, labor market, text vectorization.
