This paper presents a rule-based reasoning model for adaptive learning. The model was tested through a pilot implementation on a course at the Faculty of Economics and Administration at the University of Pardubice in Czechia. The research aimed to design a flexible system that analyses student performance and provides personalised recommendations to enhance learning outcomes.

The model was developed as a template into which data from any course can be uploaded, provided it follows a defined structure. In the first year of the pilot, the system analysed student performance and identified key problem areas. Based on these findings, targeted teaching adjustments were implemented in the second year, including supplementary materials or extended practice in problematic topics.

The results demonstrate a measurable effect: under identical course requirements, the overall pass rate for the course increased from 77.45% to 81.10%. These findings confirm that a rule-based reasoning approach can effectively support adaptive teaching, addressing student weaknesses in a structured and scalable manner.

The model is currently being refined based on pilot results and is prepared for application to additional courses. A particularly important direction is its planned use in blended study formats, which will further extend the model’s applicability and impact on personalised learning pathways in higher education.

Keywords: Adaptive learning, Rule-based reasoning, learning analytics.