<![CDATA[The rapid growth of online educational platforms has increased the demand for intelligent recommendation systems that can personalize learning content to match individual learner needs. However, traditional methods such as Content-Based Filtering (CBF) and Collaborative Filtering (CF) often struggle with issues like data sparsity, limited adaptability, and cold-start problems. This study aims to develop a personalized recommendation system for online educational content by integrating Singular Value Decomposition (SVD) with an adaptive feedback loop to improve recommendation relevance and learner engagement. The proposed machine learning-based method captures latent user-item interactions and dynamically updates recommendations based on real-time user feedback. Experimental evaluation using a dataset of simulated learner interactions demonstrates that the proposed model significantly outperforms baseline methods, achieving higher scores in Precision (0.57), Recall (0.53), F1-Score (0.55), Mean Reciprocal Rank (MRR: 0.52), and Engagement Rate (72.1%). These results suggest that combining matrix factorization with adaptive learning can substantially enhance the performance of educational recommender systems, leading to more accurate, timely, and engaging content delivery.]]>
Copyrights © 2025
