Graph Contextualized Self-Attention Network for Session-based Recommendation

By Chengfeng Xu, Pengpeng Zhao, Yanchi Liu, Victor S. Sheng, Jiajie Xu, Fuzhen Zhuang, Junhua Fang, Xiaofang Zhou. International Joint Conference on Artificial Intelligence, 2019

703 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/38b06821977b5024dd51145bbcfdd0eac42d439c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Hongzhi Yin
2. Wayne Xin Zhao
3. Junliang Yu
4. Xin Xia
5. Ji-rong Wen
6. Yupeng Hou
7. Xiangnan He
8. Yong Li

In Information Systems

1. Hongzhi Yin
2. Wayne Xin Zhao
3. Junliang Yu
4. Xin Xia
5. Yupeng Hou
6. Ji-rong Wen
7. Xiangnan He
8. Fei Sun

Citations in the corpus

• Global Context Enhanced Graph Neural Networks for Session-based Recommendation (2020, methodology): “Following the success of SR-GNN, some variants are also proposed for SBR, such as GC-SAN [25].”
• RecBole: Towards a Unified, Comprehensive and Efficient Framework for Recommendation Algorithms (2020, neutral): “GCSAN [78] IJCAI 2019 DIGINETICA, Retailrocket”
• A Survey of Graph Neural Networks for Recommender Systems: Challenges, Methods, and Directions (2021, methodology): “In session-based recommendation, most existing works [145, 187, 203, 215, 224] model the session data with a directed graph to capture the item transition pattern.”
• Handling Information Loss of Graph Neural Networks for Session-based Recommendation (2020, methodology): “Based on this work, Xu et al. [25] proposed a method that uses a GGNN to extract local context information and a self-attention network (SAN) to capture global dependencies between distant positions.”
• Graph Neural Networks in Recommender Systems: A Survey (2020, background): “GC-SAN [72] IJCAI 2019 √ Sequence graph variant of GGNN (in & out adjacent matrice)”
• Contrastive Learning for Sequential Recommendation (2022, background): “To capture both global and local structure, Xu et al. [47] combine self-attention with GNN and achieve promising results for the session-based recommendation.”
• Self-Supervised Hypergraph Convolutional Networks for Session-based Recommendation (2020, methodology): “…et al. 2019) has been reported in many areas including SBR. Unlike the RNNs-based recommendation method, the GNNs-based approaches (Wu et al. 2019b; Xu et al. 2019; Qiu et al. 2020b) model session-based data as directed subgraphs and item transitions as pairwise relations, which slightly relaxes…”
• Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation (2023, methodology): “CNN [4], [5], GNN [6], [7], and Transformer [2], [8].”
• Filter-enhanced MLP is All You Need for Sequential Recommendation (2022, methodology): “With the development of the neural networks, Hidasi et al. [17] introduced GRU to capture sequential patterns, and a surge of works leverage other neural network architectures for sequential recommendation, e.g., CNN [53], GNN [52, 60] and Transformer [22, 50].”
• AMER: Automatic Behavior Modeling and Interaction Exploration in Recommender System (2020, methodology): “SRGNN is the classical implementation to use GNN on recommendation, which represents the GNN-based recommender system such as GCSAN [77], GraphRec [15] and PinSage [78].”
• Sequential Recommendation with Graph Neural Networks (2021, background): “Recently, graph-based models [21, 31, 33, 35] achieve successes on this task.”
• A Survey on Session-based Recommender Systems (2019, methodology): “Personalized Ranking Metric Embedding (PRME) [28] O, ST, NA, SL MC, ME POI Graph Contextualized Self-Attention Network (GC-SAN) [141] O, ST, A, SL GNN, ATT Item”