Information technology for supporting learning trajectories based on recurrent neural networks and a situation-scenario approach

Abstract

The relevance of this study is determined by the need to improve the effectiveness of supporting students’ learning trajectories in distance learning systems through intelligent analysis of accumulated educational data. Modern educational platforms provide substantial amounts of information on student performance and activity; however, academic performance prediction alone does not determine subsequent pedagogical actions. The research problem lies in the need to move from obtaining predictive indicators to their situational interpretation and the formation of adaptive support scenarios. The aim of the study is to develop an information technology for supporting learning trajectories based on the integration of recurrent neural networks and a situation-scenario approach. The research employs methods of neural network forecasting of time series, situational analysis, production rules, threshold-based decision-making mechanisms, and experimental modeling. A multi-level architecture of the proposed technology is developed, encompassing data collection and preparation, forecasting, situational interpretation, scenario generation, reporting, and visualization. The RNN model provides predictions of the final grade and the probability of successful course completion, while the situation-scenario module determines the state of a learning trajectory and generates an appropriate set of pedagogical actions. A software prototype integrated with LMS Moodle was developed and tested using model data from the course Fundamentals of Programming. The results confirmed the operability of the complete technological cycle and demonstrated the possibility of generating differentiated support scenarios based on predicted outcomes and indicators of students’ learning activity. The proposed technology creates prerequisites for the timely detection of risks, adaptive support of learning trajectories, and informed pedagogical decision-making in distance education.