Abstract

Introduction. Professional education in the context of individual educational trajectories (IET) meets the needs of both students themselves and the labour market due to the relevance of the content, flexibility of the educational process and learning technologies. However, in the context of digitalisation, IET support, including their planning and subsequent management of learning, entails the emergence of new requirements for information, analytical and methodological support of information systems designed to manage the educational process of the university. The problem of this study is determined by the contradiction between the intensive growth (natural for digitalisation) in the volume and variety of types of collected data, which can and should be used to support IET. In addition, there is also a lack of adequate analytical tools in educational information management systems.Aim. The present research aimed to study and test the digitalisation methodology for IET support, based on the application of the concept of explainable artificial intelligence for analysing student digital footprint data, the content of documents regulating the educational process, as well as labour market demands.Research methodology and methods. As a theoretical basis for the study, the authors relied on the principles of explainable artificial intelligence and their application to the interpretation of data from the educational process and the prediction of educational outcomes. The methods of intellectual analysis of texts in natural language were employed for preliminary processing of source documents. To predict educational outcomes, the authors used clustering, classification and regression models created through applying machine learning methods.Results. The authors developed and studied predictive models with the subsequent formation of recommendations for the tasks of choosing an educational programme by applicants, choosing an elective discipline, forming a team for a group project and employment in accordance with professional competencies. The developed computer program automatically generates objective and explainable recommendations based on expert knowledge and predicting results. The algorithm for constructing recommendations is divided into stages and provides for variability in decision making.Scientific novelty. The authors proposed a methodology for digital support of IET, corresponding to the principles of explainable artificial intelligence, i.e. machine learning models predict educational outcomes, and a special algorithm automatically generates personalised recommendations based on the results of the analysis of data on the educational process. The developed approach confirmed its effectiveness in testing on the example of bachelor’s and master’s degree programmes in the field of computer science, information technology and information security.Practical significance. A preliminary analysis of significant volumes of initial data made it possible to obtain objective information about the data quality, including the content and structure of documents presented in various university information systems. Based on the the oretical results of the research, the authors developed a recommendation system. It included special services for students, teaching staff, tutors, and administrators, providing visual and user-oriented predictive results and recommendations. Testing of services at the Institute of Mathematics and Computer Science of University of Tyumen confirmed the feasibility of developing the functionality of the university information systems in the direction of collecting and analysing data from a student’s digital footprint and the relevance of this analysis results both by subjects of the educational process and by the labour market.

Highlights

  • Professional education in the context of individual educational trajectories (IET) meets the needs of both students themselves and the labour market due to the relevance of the content, flexibility of the educational process and learning technologies

  • As a theoretical basis for the study, the authors relied on the principles of explainable artificial intelligence and their application to the interpretation of data from the educational process and the prediction of educational outcomes

  • The authors used clustering, classification and regression models created through applying machine learning methods

Read more

Summary

ИНФОРМАЦИОННЫЕ ТЕХНОЛОГИИ

Цель статьи – исследование и апробация методологии цифровизации сопровождения ИОТ, основанной на применении концепции объяснимого искусственного интеллекта для анализа данных цифрового следа студента, содержания документов, регламентирующих образовательный процесс, а также запросов рынка труда. Теоретическую основу исследования составляет применение принципов объяснимого искусственного интеллекта к интерпретации данных об образовательном процессе и прогнозированию его результатов для выработки управляющих решений. Предложена методология цифровизации сопровождения ИОТ, отвечающая принципам объяснимого искусственного интеллекта, когда по результатам анализа данных об образовательном процессе выполняется прогноз образовательных результатов и автоматически формируются персонализированные рекомендации. Тестирование сервисов в Институте математики и компьютерных наук Тюменского государственного университета подтвердило целесообразность развития функциональных возможностей информационных систем вуза в направлении сбора и анализа данных цифрового следа студента и востребованность результатов этого анализа как субъектами образовательного процесса, так и представителями рынка труда. В. Сопровождение индивидуальных образовательных траекторий на основе концепции объяснимого искусственного интеллекта // Образование и наука. SUPPORT OF INDIVIDUAL EDUCATIONAL TRAJECTORIES BASED ON THE CONCEPT OF EXPLAINABLE ARTIFICIAL

Results
Обзор литературы
Цифровизация в образовании и объяснимый искусственный интеллект
Elements of the structure
Результаты исследования
Обсуждение результатов
Список использованных источников
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call