Knowledge Enhanced Multi-intent Transformer Network for Recommendation

TL;DR

The paper tackles the challenge of incorporating knowledge graphs into recommender systems by addressing users' multiple intents and knowledge noise. It proposes KGTN, a two-module framework that (i) models global intents via a Graph Transformer to produce intent-aware user/item representations and (ii) denoises KG through intent-guided knowledge sampling and a local-global contrastive objective. The approach yields clear offline gains across three datasets and significant online improvements in Alibaba’s production system, demonstrating both accuracy and robustness benefits. This work advances KG-enhanced recommendations by integrating global intent signals with noise-robust denoising, improving interpretability and practicality for large-scale deployments.

Abstract

Incorporating Knowledge Graphs into Recommendation has attracted growing attention in industry, due to the great potential of KG in providing abundant supplementary information and interpretability for the underlying models. However, simply integrating KG into recommendation usually brings in negative feedback in industry, due to the ignorance of the following two factors: i) users' multiple intents, which involve diverse nodes in KG. For example, in e-commerce scenarios, users may exhibit preferences for specific styles, brands, or colors. ii) knowledge noise, which is a prevalent issue in Knowledge Enhanced Recommendation (KGR) and even more severe in industry scenarios. The irrelevant knowledge properties of items may result in inferior model performance compared to approaches that do not incorporate knowledge. To tackle these challenges, we propose a novel approach named Knowledge Enhanced Multi-intent Transformer Network for Recommendation (KGTN), comprising two primary modules: Global Intents Modeling with Graph Transformer, and Knowledge Contrastive Denoising under Intents. Specifically, Global Intents with Graph Transformer focuses on capturing learnable user intents, by incorporating global signals from user-item-relation-entity interactions with a graph transformer, meanwhile learning intent-aware user/item representations. Knowledge Contrastive Denoising under Intents is dedicated to learning precise and robust representations. It leverages intent-aware representations to sample relevant knowledge, and proposes a local-global contrastive mechanism to enhance noise-irrelevant representation learning. Extensive experiments conducted on benchmark datasets show the superior performance of our proposed method over the state-of-the-arts. And online A/B testing results on Alibaba large-scale industrial recommendation platform also indicate the real-scenario effectiveness of KGTN.

Figures (6)

• Figure 1: (a) A simple case for illustrating multiple user intents with global information; (b)Performance comparison.
• Figure 2: Overall framework illustration of the proposed KGTN model. Best viewed in color.
• Figure 3: Effect of ablation study.
• Figure 4: Impact of intent number $K$.
• Figure 5: Impact of contrastive loss ratio $\alpha$.