Models and methods of e-learning individualization in the context of ontological approach

The article explores the issues of e-learning individualization (EE) as a set of processes for creating, developing, using and utilizing digital content and EE data, describes ontological models and methods for individualizing digital education. The technological stack for building and implementing individual learning paths is considered, as well as examples of existing systems that fully or partially use the specified technology stack. To efficiently process educational materials and data generated by learning management systems, a three-level architecture is proposed that allows semantic annotation and selection of concepts of varying degrees of abstraction in the data layers. These layers include: high-level abstractions of modeling, general concepts of educational materials and the educational process, specific concepts for access and integration of data from the EE system in terms of the subject area. For the first time, it was proposed to use semantic models as the formal basis for an individualized EE, including the vector representations of knowledge graphs, which, on the one hand, allows efficient processing of large and complex data structures, and on the other hand, has the flexibility and expressiveness of the ontological approach. The main aspects related to individualization in EE systems were sequentially considered, including technological aspects and existing ontologies for e-learning, individual trajectory modeling, semantic annotation of educational materials, methods for assessing knowledge in individualized learning, as well as ontological simulation of the cognitive profile of the student.