<![CDATA[The rapid growth of e-commerce platforms has led to an overwhelming number of product choices, creating challenges for users in identifying items that match their preferences. Recommendation systems have become essential tools to address this issue; however, traditional approaches such as Collaborative Filtering and Content-Based Filtering suffer from limitations including data sparsity, cold-start problems, and limited recommendation diversity. This study proposes a Hybrid Filtering-based product recommendation system that integrates both Collaborative Filtering and Content-Based Filtering techniques to overcome these challenges. The proposed model utilizes user-item interaction data and product metadata to generate personalized recommendations through a hybrid approach, combining algorithms such as K-Nearest Neighbors (KNN), Matrix Factorization, Term Frequency–Inverse Document Frequency (TF-IDF), and cosine similarity. The system is evaluated using multiple performance metrics, including accuracy (precision, recall, and F1-score), ranking quality (Mean Average Precision and Normalized Discounted Cumulative Gain), and prediction error (Root Mean Square Error and Mean Absolute Error). The results demonstrate that the Hybrid Filtering model outperforms individual methods in all evaluation aspects. It achieves higher accuracy, better ranking performance, lower prediction error, and greater diversity in recommendations. These findings indicate that the hybrid approach effectively addresses the limitations of traditional recommendation systems and provides more reliable and personalized recommendations. In conclusion, this research confirms that Hybrid Filtering is a robust and efficient method for improving the performance of product recommendation systems. The proposed model has significant practical implications for e-commerce platforms, as it enhances user experience, increases engagement, and supports better decision-making processes.]]>
Copyrights © 2026
