Exploiting Cross-Session Information for Session-based Recommendation with Graph Neural Networks

TL;DR

This paper introduces a graph neural network approach for session-based recommendation that captures complex item dependencies within sessions and leverages cross-session information to improve recommendation accuracy.

Contribution

It proposes a novel Full Graph Neural Network and a Broadly Connected Session graph to incorporate cross-session data, enhancing session embedding quality.

Findings

Outperforms state-of-the-art models on Yoochoose and Diginetica datasets.

Effectively captures complex item dependencies within sessions.

Utilizes cross-session information to improve recommendation performance.

Abstract

Different from the traditional recommender system, the session-based recommender system introduces the concept of the session, i.e., a sequence of interactions between a user and multiple items within a period, to preserve the user's recent interest. The existing work on the session-based recommender system mainly relies on mining sequential patterns within individual sessions, which are not expressive enough to capture more complicated dependency relationships among items. In addition, it does not consider the cross-session information due to the anonymity of the session data, where the linkage between different sessions is prevented. In this paper, we solve these problems with the graph neural networks technique. First, each session is represented as a graph rather than a linear sequence structure, based on which a novel Full Graph Neural Network (FGNN) is proposed to learn complicated item dependency. To exploit and incorporate cross-session information in the individual session's representation learning, we further construct a Broadly Connected Session (BCS) graph to link different sessions and a novel Mask-Readout function to improve session embedding based on the BCS graph. Extensive experiments have been conducted on two e-commerce benchmark datasets, i.e., Yoochoose and Diginetica, and the experimental results demonstrate the superiority of our proposal through comparisons with state-of-the-art session-based recommender models.