Artificial Intelligence-Based Personalized Learning Recommendation System for Secondary School Students

Abstract

The increasing adoption of Artificial Intelligence (AI) in education has created new opportunities for addressing the limitations of traditional teaching approaches, which often fail to accommodate the diverse learning needs, abilities, and academic backgrounds of students. Personalized learning systems have emerged as a promising solution for improving student engagement and academic outcomes through data-driven interventions and adaptive content delivery. This study presents the development of an AI-based personalized learning recommendation system for secondary school students. The research integrates supervised machine learning classification models with collaborative filtering techniques to predict students’ academic performance and generate adaptive learning recommendations. Using the UCI Student Performance Dataset, multiple classification algorithms, including Logistic Regression, Random Forest, XGBoost, Support Vector Machine (SVM), and K-Nearest Neighbors (KNN), were evaluated. Random Forest achieved the highest predictive performance by modeling complex nonlinear relationships in educational data, demonstrating its effectiveness in capturing complex nonlinear relationships within educational data. A dual-layer recommendation framework combining rule-based mapping and Singular Value Decomposition (SVD) was implemented to enhance personalization. The recommendation module also generated meaningful personalized learning suggestions, although its effectiveness was influenced by classification errors, particularly in distinguishing at-risk and average-performing students.  The findings confirm that ensemble learning methods are effective for educational data analytics and that AI-driven recommendation systems can support early intervention, targeted remediation, and improved learning outcomes in secondary education.