Rethinking the Item Order in Session-based Recommendation with Graph Neural Networks

By Ruihong Qiu, Jingjing Li, Zi-Liang Huang, Hongzhi Yin. International Conference on Information and Knowledge Management, 2019

417 citations recorded by the source corpus.

Predicting a user's preference in a short anonymous interaction session instead of long-term history is a challenging problem in the real-life session-based recommendation, e.g., e-commerce and media stream. Recent research of the session-based recommender system mainly focuses on sequential patterns by utilizing the attention mechanism, which is straightforward for the session's natural sequence sorted by time. However, the user's preference is much more complicated than a solely consecutive time pattern in the transition of item choices. In this paper, therefore, we study the item transition pattern by constructing a session graph and propose a novel model which collaboratively considers the sequence order and the latent order in the session graph for a session-based recommender system. We formulate the next item recommendation within the session as a graph classification problem. Specifically, we propose a weighted attention graph layer and a Readout function to learn embeddings of items and sessions for the next item recommendation. Extensive experiments have been conducted on two benchmark E-commerce datasets, Yoochoose and Diginetica, and the experimental results show that our model outperforms other state-of-the-art methods.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/b0d941cfa0a3c43703e5222221addb1ad0f9e68d

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. Junliang Yu
3. Xin Xia
4. Zi-Liang Huang
5. Yong Li
6. Chen Gao
7. Xiangnan He
8. Li-zhen Cui

In Information Systems

1. Hongzhi Yin
2. Junliang Yu
3. Xin Xia
4. Zi-Liang Huang
5. Yong Li
6. Chen Gao
7. Xiangnan He
8. Li-zhen Cui

Citations in the corpus

• Contrastive Learning for Representation Degeneration Problem in Sequential Recommendation (2021, methodology): “Recent graphbased methods, FGNN [30, 32], GAG [34] and PosRec [29], achieve improved performance due to the graph modeling in sequence.”
• Handling Information Loss of Graph Neural Networks for Session-based Recommendation (2020, methodology): “6Note that [11, 16, 17, 23] used different metric names for HR@20 (e.”
• Global Context Enhanced Graph Neural Networks for Session-based Recommendation (2020, methodology): “FGNN12 [9]: It is recently proposed by designing a weighted attention graph layer to learn items embeddings, and the sessions for the next item recommendation are learnt by a graph level feature extractor.”
• A Survey on Session-based Recommender Systems (2019, methodology): “GAT-based SBRSs mainly utilize attention mechanism to attentively integrate the information from the neighbourhood nodes of node ni in a session graph to update its hidden state in each attention layer [82]:”
• Self-Supervised Hypergraph Convolutional Networks for Session-based Recommendation (2020, background): “FGNN (Qiu et al. 2019) constructs a session graph to learn item transition pattern and rethinks the sequence order of items in SBR. GCE-GNN (Wang et al. 2020c) conduct graph convolution on both the single session graph and the global session graph to learn session-level and global-level embed-dings.”
• Graph Neural Networks in Recommender Systems: A Survey (2020, background): “However, the users’ preferences are much more complicated than a solely consecutive time pattern, and the inherent order is not the complete randomness by the self-attention [42].”
• Sequential Recommendation with Graph Neural Networks (2021, background): “Recently, graph-based models [21, 31, 33, 35] achieve successes on this task.”
• Contrastive Learning for Sequential Recommendation (2022, methodology): “Qiu et al. [35] utilize the attention mechanism as the aggregation function to differentiate the influence of different neighbors instead of mean pooling.”
• 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.”