Intelligent adaptive learning system based on neural networks for personalization of educational trajectories of Russian university students

One of the ways to improve the quality and efficiency of higher education in Russia is to personalize learning, to find ways to create an individual educational trajectory for each student. Traditional teaching methods do not consider the individual characteristics of students, while the development of digital technologies, including artificial intelligence technologies, creates new ways to transform the higher education system. The article presents a tool for personalizing the educational trajectories of students in Russian universities — an intelligent adaptive learning system based on neural networks. During the research described in the article, the optimal neural network architecture was determined (using multilayer perсeptron, convolutional and recurrent neural networks). The created adaptive system was integrated into the infrastructure of five Russian universities, where an experiment on its application involving 1000 students was conducted. Then, the impact of the intelligent adaptive learning system on students’ academic performance and engagement as well as on teaching efficiency was evaluated. The evaluation methods were statistical analysis using Student’s t-test and analysis of variance, student and faculty surveys. The results of the experimental application of the intelligent adaptive learning system demonstrate an accuracy of predicting the performance of the educational process up to 92 %, an increase the academic performance of students in the experimental group by 18 %, an increase of their involvement in the learning process by 25 %, as well as an increase in the efficiency of teachers’ work by 15 % with a decrease in time costs by 20 %. The research presented in this article demonstrates a significant potential of intelligent adaptive learning systems to improve the quality of higher education in Russia. The results can be used for optimization of educational programs and teaching methods. Further research is needed on the long-term impact of the presented adaptive learning system on the career trajectories of graduates and the labor market demand for the graduates.