Knowledge Graph Self-Supervised Rationalization for Recommendation

TL;DR

This paper presents KGRec, a self-supervised method for knowledge-aware recommendation that identifies and highlights useful knowledge connections through rationalization, improving recommendation accuracy.

Contribution

The paper introduces a novel attentive rationalization mechanism and combines generative and contrastive self-supervised tasks for knowledge graph-based recommendation.

Findings

KGRec outperforms state-of-the-art methods on three datasets.

The rationalization mechanism effectively highlights important knowledge triplets.

Contrastive learning improves robustness against noisy knowledge edges.

Abstract

In this paper, we introduce a new self-supervised rationalization method, called KGRec, for knowledge-aware recommender systems. To effectively identify informative knowledge connections, we propose an attentive knowledge rationalization mechanism that generates rational scores for knowledge triplets. With these scores, KGRec integrates generative and contrastive self-supervised tasks for recommendation through rational masking. To highlight rationales in the knowledge graph, we design a novel generative task in the form of masking-reconstructing. By masking important knowledge with high rational scores, KGRec is trained to rebuild and highlight useful knowledge connections that serve as rationales. To further rationalize the effect of collaborative interactions on knowledge graph learning, we introduce a contrastive learning task that aligns signals from knowledge and user-item interaction views. To ensure noise-resistant contrasting, potential noisy edges in both graphs judged by the rational scores are masked. Extensive experiments on three real-world datasets demonstrate that KGRec outperforms state-of-the-art methods. We also provide the implementation codes for our approach at https://github.com/HKUDS/KGRec.