VRKG4Rec: Virtual Relational Knowledge Graphs for Recommendation

TL;DR

VRKG4Rec introduces a novel recommendation model that constructs virtual relational graphs and employs a local weighted smoothing mechanism to effectively incorporate relational knowledge, improving recommendation accuracy.

Contribution

The paper proposes VRKG4Rec, which constructs virtual relational graphs and uses a parameter-free smoothing mechanism for better item and user representations in recommendation systems.

Findings

VRKG4Rec outperforms state-of-the-art methods on two datasets.

The virtual relational graphs effectively encode relational knowledge.

The local weighted smoothing mechanism improves embedding quality.

Abstract

Incorporating knowledge graph as side information has become a new trend in recommendation systems. Recent studies regard items as entities of a knowledge graph and leverage graph neural networks to assist item encoding, yet by considering each relation type individually. However, relation types are often too many and sometimes one relation type involves too few entities. We argue that it is not efficient nor effective to use every relation type for item encoding. In this paper, we propose a VRKG4Rec model (Virtual Relational Knowledge Graphs for Recommendation), which explicitly distinguish the influence of different relations for item representation learning. We first construct virtual relational graphs (VRKGs) by an unsupervised learning scheme. We also design a local weighted smoothing (LWS) mechanism for encoding nodes, which iteratively updates a node embedding only depending on the embedding of its own and its neighbors, but involve no additional training parameters. We also employ the LWS mechanism on a user-item bipartite graph for user representation learning, which utilizes encodings of items with relational knowledge to help training representations of users. Experiment results on two public datasets validate that our VRKG4Rec model outperforms the state-of-the-art methods. The implementations are available at https://github.com/lulu0913/VRKG4Rec.