Multi-hop Multi-view Memory Transformer for Session-based Recommendation

Abstract

A S ession- B ased R ecommendation (SBR) seeks to predict users’ future item preferences by analyzing their interactions with previously clicked items. In recent approaches, G raph N eural N etworks (GNNs) have been commonly applied to capture item relations within a session to infer user intentions. However, these GNN-based methods typically struggle with feature ambiguity between the sequential session information and the item conversion within an item graph, which may impede the model's ability to accurately infer user intentions. In this paper, we propose a novel M ulti-hop M ulti-view M emory T ransformer ( \(\rm{M^{3}T}\) ) to effectively integrate the sequence-view information and relation conversion (graph-view information) of items in a session. First, we propose a M ulti-view M emory T ransformer ( \(\rm{M^{2}T}\) ) module to concurrently obtain multi-view information of items. Then, a set of trainable memory matrices are employed to store sharable item features, which mitigates cross-view item feature ambiguity. To comprehensively capture latent user intentions, a M ulti-hop \(\rm{M^{2}T}\) ( \(\rm{M^{3}T}\) ) framework is designed to integrate user intentions across different hops of an item graph. Specifically, a k-order power method is proposed to manage the item graph to alleviate the over-smoothing problem when obtaining high-order relations of items. Extensive experiments conducted on three real-world datasets demonstrate the superiority of our method.