Graph Collaborative Signals Denoising and Augmentation for Recommendation

TL;DR

This paper introduces a novel graph adjacency matrix for recommendation systems that incorporates user-user and item-item correlations, along with a balanced user-item interaction matrix, to improve recommendation accuracy especially for users with varying interaction levels.

Contribution

The work proposes a new graph adjacency matrix and interaction matrix that enhance neighbor quality and diversity, improving recommendation performance for all users.

Findings

Enhanced adjacency matrix improves recommendation accuracy.

Inclusion of user-user and item-item correlations benefits users with different interaction levels.

Balanced interaction matrix reduces noise and improves neighbor relevance.

Abstract

Graph collaborative filtering (GCF) is a popular technique for capturing high-order collaborative signals in recommendation systems. However, GCF's bipartite adjacency matrix, which defines the neighbors being aggregated based on user-item interactions, can be noisy for users/items with abundant interactions and insufficient for users/items with scarce interactions. Additionally, the adjacency matrix ignores user-user and item-item correlations, which can limit the scope of beneficial neighbors being aggregated. In this work, we propose a new graph adjacency matrix that incorporates user-user and item-item correlations, as well as a properly designed user-item interaction matrix that balances the number of interactions across all users. To achieve this, we pre-train a graph-based recommendation method to obtain users/items embeddings, and then enhance the user-item interaction matrix via top-K sampling. We also augment the symmetric user-user and item-item correlation components to the adjacency matrix. Our experiments demonstrate that the enhanced user-item interaction matrix with improved neighbors and lower density leads to significant benefits in graph-based recommendation. Moreover, we show that the inclusion of user-user and item-item correlations can improve recommendations for users with both abundant and insufficient interactions. The code is in \url{https://github.com/zfan20/GraphDA}.