M.K. Murodov

Annotation:

The article presents and substantiates the mathematical and algorithmic models of an intelligent platform for assessing and teaching programming. The proposed approach models a student’s code as a discrete structure and identifies and classifies errors based on componentwise and structural analysis. The main classes of errors are distinguished as syntactic, logical, structural, and parametric (or efficiency-related) errors. Decision-making is carried out on the basis of a set of features and parameters extracted from the code.The feedback mechanism and personalized recommendations are implemented as an iterative process, which allows refining recommendations at each iteration by taking into account the learning history and the dynamics of the student’s progress. To evaluate learning effectiveness. Experimental validation of the proposed method, performed with 42 student participants, confirmed its significant effectiveness.The results of the theoretical analysis show that the integration of error classification, structural code analysis, and an adaptive recommendation mechanism can improve the quality of programming education and provide a scientific foundation for the development of intelligent learning environments.

Keywords: programming, intelligent learning environment, error classification, structural code analysis, mathematical modeling, adaptive algorithms, feedback mechanism, personalized recommendations.