Coarse-to-Fine Sparse Sequential Recommendation

Abstract

Sequential recommendation aims to model dynamic user behavior from historical interactions. Self-attentive methods have proven effective at capturing short-term dynamics and long-term preferences. Despite their success, these approaches still struggle to model sparse data, on which they struggle to learn high-quality item representations. We propose to model user dynamics from shopping intents and interacted items simultaneously. The learned intents are coarse-grained and work as prior knowledge for item recommendation. To this end, we present a coarse-to-fine self-attention framework, namely CaFe, which explicitly learns coarse-grained and fine-grained sequential dynamics. Specifically, CaFe first learns intents from coarse-grained sequences which are dense and hence provide high-quality user intent representations. Then, CaFe fuses intent representations into item encoder outputs to obtain improved item representations. Finally, we infer recommended items based on representations of items and corresponding intents. Experiments on sparse datasets show that CaFe outperforms state-of-the-art self-attentive recommenders by 44.03% NDCG@5 on average.

Personalized Behavior-Aware Transformer for Multi-Behavior Sequential Recommendation

“…Early work usually models the sequential dependencies with the Markov Chain assumption [24]. With the advance in deep learning, Recurrent Neural Networks (RNN) based [10,11,20,35], Convolutional Neural Networks (CNN) based [28], Graph Neural Networks (GNN) based [26,37,51], and attention based [13,16,27,36] models have been adopted to explore dynamic user interests that are hidden underneath behavior sequences. Recently, contrastive learning based models [19,23,41] are introduced to extract meaningful user patterns by deriving self-supervision signals.…”

“…), position-enhanced behavior-aware fusion (line 10-12), get attention scores (line[13][14], and attentive aggregation (line[16][17][18][19][20].…”