Implementation of a Machine Learning-Based Adaptive Learning System for Personalizing Student Learning Paths

Abstract

This study addresses the problem of how to design and implement an adaptive learning system to support students in the Modeling and Simulation course, where diverse backgrounds and learning difficulties often hinder the achievement of optimal outcomes. The novelty of this research lies in the integration of a machine learning approach, specifically a Decision Tree Classifier, to personalize students' learning paths based on their pretest performance. The research adopts a Research and Development (R&D) design using the ADDIE model, which consists of five stages: Analysis, Design, Development, Implementation, and Evaluation. Data collection methods included pretest–posttest, questionnaires, and interviews, while data were analyzed both quantitatively to measure learning improvement and qualitatively to capture students' learning experiences. The main results show that the adaptive system effectively categorized students into three distinct learning paths remedial, standard, and enrichment each with tailored content to match students' competencies. The evaluation stage demonstrated an increase in learning outcomes and positive student feedback regarding motivation and satisfaction. These findings indicate that the adaptive learning system not only improved cognitive achievement but also enhanced students' engagement, offering a practical contribution to the development of technology-assisted personalized learning in higher education.