WITH MACHINE LEARNING BASED ON EDUCATIONAL DATASTUDENT LEARNING PREDICTION MODEL

Abstract

This study aims to develop a model for predicting students’ knowledge levels using machine learning methods based on educational data. In digital learning environments, the growing volume of educational data provides significant opportunities for personalizing the learning process and improving academic outcomes; however, these opportunities are not sufficiently utilized in practice. Therefore, the development of intelligent models capable of predicting student learning performance in advance remains a pressing research issue. Within the research methodology, a dataset reflecting the academic activities of students in a higher education institution was constructed and subjected to initial preprocessing. Predictive models were developed using linear regression, decision trees, random forest, and artificial neural network algorithms. Model performance was evaluated and compared using accuracy and error metrics. The results indicate that ensemble-based models achieve higher accuracy in predicting students’ knowledge levels. The developed model can be effectively applied to early identification of at-risk students, the design of adaptive learning strategies, and the improvement of educational quality. The findings provide a scientific and practical foundation for integrating artificial intelligence technologies into the education system and for supporting data-driven pedagogical decision-making