Recent Papers

Recommender systems play a crucial role in personalizing user experiences across various domains, including e-commerce, entertainment, and online learning. Matrix factorization has emerged as a powerful technique for uncovering latent patterns in user-item interactions, enabling accurate and scalable recommendation models. This paper explores the fundamentals of matrix factorization, including Singular Value Decomposition (SVD), Non-negative Matrix Factorization (NMF), and Alternating Least Squares (ALS), while highlighting their applications in collaborative filtering. We also discuss key challenges such as sparsity, cold start problems, and scalability, along with recent advancements incorporating deep learning and hybrid approaches. By leveraging matrix factorization, businesses and platforms can enhance user engagement through more personalized and relevant recommendations. This structure can reduce the model's calculation time by extracting the nonlinear features of both item and user latent characteristics at the same time. Furthermore, the model incorporates slide-attention, an enhanced attention technique. The algorithm handles the interaction problem between the item's latent characteristics and the user's various dimensions by using the sliding query approach to draw the user's attention to the item's latent features.

Copyright © 2025 J.Vidhyajanani, . This is an open access article distributed under the Creative Commons Attribution License.