Sequential Recommendation with Graph Neural Networks

TL;DR

This paper introduces SURGE, a graph neural network model that improves sequential recommendation by explicitly modeling core user interests from noisy, long-term behavior sequences, achieving significant performance gains.

Contribution

The paper proposes SURGE, a novel GNN-based approach that reconstructs user behavior sequences into interest graphs and applies cluster-aware graph convolutions to better capture dynamic user preferences.

Findings

SURGE outperforms state-of-the-art methods on multiple datasets.

The model effectively captures long-term user interests.

Sequence length studies show robustness and efficiency.

Abstract

Sequential recommendation aims to leverage users' historical behaviors to predict their next interaction. Existing works have not yet addressed two main challenges in sequential recommendation. First, user behaviors in their rich historical sequences are often implicit and noisy preference signals, they cannot sufficiently reflect users' actual preferences. In addition, users' dynamic preferences often change rapidly over time, and hence it is difficult to capture user patterns in their historical sequences. In this work, we propose a graph neural network model called SURGE (short for SeqUential Recommendation with Graph neural nEtworks) to address these two issues. Specifically, SURGE integrates different types of preferences in long-term user behaviors into clusters in the graph by re-constructing loose item sequences into tight item-item interest graphs based on metric learning. This helps explicitly distinguish users' core interests, by forming dense clusters in the interest graph. Then, we perform cluster-aware and query-aware graph convolutional propagation and graph pooling on the constructed graph. It dynamically fuses and extracts users' current activated core interests from noisy user behavior sequences. We conduct extensive experiments on both public and proprietary industrial datasets. Experimental results demonstrate significant performance gains of our proposed method compared to state-of-the-art methods. Further studies on sequence length confirm that our method can model long behavioral sequences effectively and efficiently.