Improvement Graph Convolution Collaborative Filtering with Weighted addition input

TL;DR

This paper introduces WiGCN, a graph-based collaborative filtering model that augments the user-item interaction graph with a weighted input graph to capture varying user influence. It constructs a bi-partite graph and a weighted references matrix, feeding two Laplacian-like inputs $\Gamma$ and $\Delta$ into a multi-layer graph neural framework with embeddings updated by $E^{(k)} = LeakyReLU[ \Gamma E^{(k-1)}W_1^{(k-1)} + \Delta \Gamma E^{(k-1)}W_2^{(k-1)} + b^{(k-1)}]$, and optimizes via Bayesian Personalized Ranking. Empirical results on Gowalla, Amazon-book, and Yelp2018 show WiGCN achieving higher precision, recall, and ndcg than GCMC, NGCF, and LightGCN in most cases, with faster loss convergence, indicating better utilization of weighted interaction signals. The approach highlights the value of incorporating a weighted trust-like input to modulate information flow in graph-based recommender systems, offering practical improvements for implicit-feedback scenarios. Code is provided at the authors’ GitHub repository for reproducibility and further exploration.

Abstract

Graph Neural Networks have been extensively applied in the field of machine learning to find features of graphs, and recommendation systems are no exception. The ratings of users on considered items can be represented by graphs which are input for many efficient models to find out the characteristics of the users and the items. From these insights, relevant items are recommended to users. However, user's decisions on the items have varying degrees of effects on different users, and this information should be learned so as not to be lost in the process of information mining. In this publication, we propose to build an additional graph showing the recommended weight of an item to a target user to improve the accuracy of GNN models. Although the users' friendships were not recorded, their correlation was still evident through the commonalities in consumption behavior. We build a model WiGCN (Weighted input GCN) to describe and experiment on well-known datasets. Conclusions will be stated after comparing our results with state-of-the-art such as GCMC, NGCF and LightGCN. The source code is also included at https://github.com/trantin84/WiGCN.

Figures (3)

• Figure 1: Bi-parties users-items graph and Graph of users' influences.
• Figure 2: The proposed framework with bi-parties graph and Weighted influence users as additional input.
• Figure 3: Test performance of each epoch of MF and NGCF.