Формирование рекомендаций по контенту для пользователей образовательных порталов на основе машинного обучения

Abstract

The paper proposes creating a scientific approach to forming content recommendations in the educational portal being developed for students, as well as for professionals who wish to improve their skills or people who are engaged in self-education. The relevance of the study lies in the fact that at present the learning process individualization is a new current trend in developing educational activities. It allows fine-tuning the system and its content to each user’s requirements and preferences. Unlike the universal approach on which classical educational portals are based, which provide training courses focused on the general public and built according to a template, the portal being developed on the learning individualization principle is focused on the student, and such a system can be additionally customized by him. To give recommendations, it is necessary to analyse user data collected in the course of his work in the system, as well as data on educational materials placed in the system, presented in the metadata form. The article describes the process of analysing data about users and educational materials to form recommendations about the educational content for the students. Two approaches to searching for similarities between the elements of the projected educational portal are compared to give content recommendations and examples show their applicability in e-learning systems