Ivan Krechetov — Head of the Laboratory of Instrumental Modelling and Learning Systems, Tomsk State University of Control Systems and Radioelectronics. E-mail: kia@2i.tusur.ru

Vladimir Romanenko — Candidate of Sciences in Technology, Associate Professor, Department of Automated Control Systems, Tomsk State University of Control Systems and Radioelectronics. E-mail: rva@2i.tusur.ru

Address: Room 607, 146 Krasnoarmeyskaya Str., 634034 Tomsk, Russian Federation.

The concept of adaptive learning emerged a few decades ago, but most theoretical findings have never been put into practice, and software solutions had no significant reach for a long time due to insufficient e-learning technology development and coverage. The recent advancements of information technology allow the elaboration of complex big data analytics and artificial intelligence solutions, in adaptive learning in particular.This article investigates exploitation of adaptive learning technology and techniques.The solutions proposed allow mapping optimal individualized learning paths for students in online courses, using the ratio of the level of knowledge at course completion to time spent on the course as an optimality criterion. A genetic algorithm is used to solve this optimization problem. A model based on the speed of forgetting was applied to extrapolate the level of retained knowledge. Practical implementation of the technology proposed involves a set of tools to expand the adaptive learning opportunities of distance learning systems and a module to operate the genetic algorithm. We developed a few options of software architecture using different technologies and programming languages and either one or two servers. The solution was tested during the design of adaptive learning courses for National University of Science and Technology MISIS (NUST MISIS) and Tomsk State University of Control Systems and Radioelectronics (TUSUR).