Based Collaborative Filtering

Graph-based collaborative filtering (GCF) aims to improve recommendation systems by leveraging the relationships between users and items represented as a graph. Current research focuses on enhancing GCF's accuracy and efficiency through techniques like advanced sampling strategies (e.g., contrastive learning and hard negative sampling), optimized graph neural networks (GNNs), and computationally efficient algorithms that avoid matrix decomposition. These advancements address challenges such as data sparsity and computational cost, leading to more accurate and faster personalized recommendations with applications across various domains.