Graph Contextualized Self-Attention Network for Session-based Recommendation

Abstract

Session-based recommendation, which aims to predict the user's immediate next action based on anonymous sessions, is a key task in many online services (e.g., e-commerce, media streaming). Recently, Self-Attention Network (SAN) has achieved significant success in various sequence modeling tasks without using either recurrent or convolutional network. However, SAN lacks local dependencies that exist over adjacent items and limits its capacity for learning contextualized representations of items in sequences. In this paper, we propose a graph contextualized self-attention model (GC-SAN), which utilizes both graph neural network and self-attention mechanism, for session-based recommendation. In GC-SAN, we dynamically construct a graph structure for session sequences and capture rich local dependencies via graph neural network (GNN). Then each session learns long-range dependencies by applying the self-attention mechanism. Finally, each session is represented as a linear combination of the global preference and the current interest of that session. Extensive experiments on two real-world datasets show that GC-SAN outperforms state-of-the-art methods consistently.

Recommendation System Based on Heterogeneous Feature: A Survey

“…In 2019, Xu et al [13] proposed a graph contextualized self-attention model (GC-SAN), which uses both GNN and self-attention mechanism, for session-based recommendations. They dynamically constructed a graph structure for session sequence and captured deep local dependencies using GNN.…”

“…Predicting the next location: A recurrent model with spatial and temporal contexts [47] https://github.com/yongqyu/STRNN A neural network approach to jointly modelling https://github.com/thunlp/JNTM A content-collaborative recommender that exploits WordNet-based user profiles for neighbourhood formation [49] https://github.com/groveco/content-engine Collaborative metric learning [52] https://github.com/changun/CollMetric [9] Pinsage 0.591 0.67 [10] STACR-GCN 0.895 [11] NGNN 0.7701 0.96 [12] GraphRec 0.9794 [13] GC-SAN 0.284 0.535 [14] KGAT 0.149 [15] GC-MC 0.905 [17] CR-CA 0.418 [18] Node2vec 0.224 0.155 0.441 [22] IDGCCF 0.976 0.882 0.876 0.976 Text-based recommendation [26] CO-Attention 0.311 0.286 [27] LSTM-CAV 0.724 0.309 [28] CNNRNN 0.532 0.416 0.698 [29] DL 0.288 [30] GRU-MTL 0.605 Behavior-based recommendation system [7] SA-UserCF 0.1571 0.082 0.171 [36] DBNCF 0.7742…”

Group Recommendation via Self-Attention and Collaborative Metric Learning Model

“…In this paper, four evaluation indicators are used to evaluate the performance of the model: precision [31], recall [31], AUC [32] and NDCG [33]. Let K be the number of recommendation services where K is set to 5,10 and 15.…”

Beyond Clicks: Modeling Multi-Relational Item Graph for Session-Based Target Behavior Prediction

“…In the domain of single-session based behavior prediction, some studies [14,22,25] adopt attention mechanism [1,28] and outperform the pioneering RNN based methods [8]. Recent advances in graph neural networks (GNN) [3,7] further boost the performance of session-based behavior prediction by modeling each sessionbased behavior sequence as a graph to achieve the state-of-the-art performance [29,30]. However, existing studies in this regard still suffer from several limitations.…”