Neural Graph Collaborative Filtering

Xiang, Wang; He, Xiangnan; Wang, Meng; Feng, Fuli; Chua, Tat-Seng

doi:10.1145/3331184.3331267

Cited by 2,337 publications

(1,708 citation statements)

References 30 publications

Supporting

Mentioning

1,534

Contrasting

Order By: Relevance

“…Gowalla: This is the check-in dataset [29] obtained from Gowalla, in which users share their locations by check-in. To ensure the quality of the dataset, we retain users and items with at least ten interactions similar to [53].…”

Section: Datasetsmentioning

confidence: 99%

Learning to Hash with Graph Neural Networks for Recommender Systems

Tan

Liu

Zhao

et al. 2020

Proceedings of the Web Conference 2020

View full text Add to dashboard Cite

Recommender systems in industry generally include two stages: recall and ranking. Recall refers to efficiently identify hundreds of candidate items that user may interest in from a large volume of item corpus, while the latter aims to output a precise ranking list using complex ranking models. Recently, graph representation learning has attracted much attention in supporting high quality candidate search at scale. Despite its effectiveness in learning embedding vectors for objects in the user-item interaction network, the computational costs to infer users' preferences in continuous embedding space are tremendous. In this work, we investigate the problem of hashing with graph neural networks (GNNs) for high quality retrieval, and propose a simple yet effective discrete representation learning framework to jointly learn continuous and discrete codes. Specifically, a deep hashing with GNNs (HashGNN) is presented, which consists of two components, a GNN encoder for learning node representations, and a hash layer for encoding representations to hash codes. The whole architecture is trained end-to-end by jointly optimizing two losses, i.e., reconstruction loss from reconstructing observed links, and ranking loss from preserving the relative ordering of hash codes. A novel discrete optimization strategy based on straight through estimator (STE) with guidance is proposed. The principal idea is to avoid gradient magnification in back-propagation of STE with continuous embedding guidance, in which we begin from learning an easier network that mimic the continuous embedding and let it evolve during the training until it finally goes back to STE. Comprehensive experiments over three publicly available and one real-world Alibaba company datasets demonstrate that our model not only can achieve comparable performance compared with its continuous counterpart but also runs multiple times faster during inference.

show abstract

Section: Datasetsmentioning

confidence: 99%

Learning to Hash with Graph Neural Networks for Recommender Systems

Tan

Liu

Zhao

et al. 2020

Proceedings of the Web Conference 2020

View full text Add to dashboard Cite

show abstract

“…whereŷ ui is the prediction score for an (u, i) interaction; f R (·) is abstracted as the interaction function with recommender parameters Θ R ; r u ∈ R d and r i ∈ R d are ID embeddings of user u and item i, respectively; d is the embedding size. Following prior studies [12,23,34], we use the pairwise BPR loss [23] as the objective function to optimize and learn the parameters Θ R . Specifically, it assumes that, for a target user, her historical items reflecting more personal interest should be assigned higher prediction scores, than that of unobserved items, as:…”

Section: Recommendermentioning

confidence: 99%

“…3.3.1 Graph Learning Module. Inspired by recent graph neural networks (GNNs) [10,17,26,34] which are powerful to generate representations for graph data, we employ GraphSage [10] on G and the user-item bipartite graph O + , to embed user, item, and KG entity nodes. In particular, at the l-th graph convolutional layers, a node e receives the information being propagated from its neighbors to update its representation, as:…”

Section: Knowledge Graph Policy Networkmentioning

confidence: 99%

Reinforced Negative Sampling over Knowledge Graph for Recommendation

Wang

Xu²,

et al. 2020

Proceedings of the Web Conference 2020

Self Cite

170

114

View full text Add to dashboard Cite

Properly handling missing data is a fundamental challenge in recommendation. Most present works perform negative sampling from unobserved data to supply the training of recommender models with negative signals. Nevertheless, existing negative sampling strategies, either static or adaptive ones, are insufficient to yield high-quality negative samples -both informative to model training and reflective of user real needs.In this work, we hypothesize that item knowledge graph (KG), which provides rich relations among items and KG entities, could be useful to infer informative and factual negative samples. Towards this end, we develop a new negative sampling model, Knowledge Graph Policy Network (KGPolicy), which works as a reinforcement learning agent to explore high-quality negatives. Specifically, by conducting our designed exploration operations, it navigates from the target positive interaction, adaptively receives knowledgeaware negative signals, and ultimately yields a potential negative item to train the recommender. We tested on a matrix factorization (MF) model equipped with KGPolicy, and it achieves significant improvements over both state-of-the-art sampling methods like DNS [39] and IRGAN [30], and KG-enhanced recommender models like KGAT [32]. Further analyses from different angles provide insights of knowledge-aware sampling. We release the codes and datasets at https://github.com/xiangwang1223/kgpolicy. CCS CONCEPTS• Information systems → Recommender systems.

show abstract

“…SpectralCF [10] designs a spectral convolutional filter to model the CF signals in user-item bipartite graphs. NGCF [11] explicitly models the high-order CF signal in the user-item bipartite graph. It designs a multi-layer graph convolutional network by constructing, then aggregating the messages over the graph.…”

Section: Related Workmentioning

confidence: 99%

“…The basket recommendation problem can thus be defined as predicting the links between basket nodes and item nodes. On top of this graph, we propose a new framework named BasConv to tackle this problem with a graph convolutional neural network (GCNN) [13,11,14]. Different from prior work, we are able to address the following two aspects:…”

Section: Introductionmentioning

confidence: 99%

BasConv: Aggregating Heterogeneous Interactions for Basket Recommendation with Graph Convolutional Neural Network

Liu¹,

Wan²,

Guo³

et al. 2020

Proceedings of the 2020 SIAM International Conference on Data Mining

View full text Add to dashboard Cite

Within-basket recommendation reduces the exploration time of users, where the user's intention of the basket matters. The intent of a shopping basket can be retrieved from both user-item collaborative filtering signals and multi-item correlations. By defining a basket entity to represent the basket intent, we can model this problem as a basket-item link prediction task in the User-Basket-Item (UBI) graph. Previous work solves the problem by leveraging user-item interactions and item-item interactions simultaneously. However, collectivity and heterogeneity characteristics are hardly investigated before. Collectivity defines the semantics of each node which should be aggregated from both directly and indirectly connected neighbors. Heterogeneity comes from multi-type interactions as well as multi-type nodes in the UBI graph. To this end, we propose a new framework named BasConv, which is based on the graph convolutional neural network. Our BasConv model has three types of aggregators specifically designed for three types of nodes. They collectively learn node embeddings from both neighborhood and high-order context. Additionally, the interactive layers in the aggregators can distinguish different types of interactions. Extensive experiments on two real-world datasets prove the effectiveness of BasConv. * work is done during internship at Walmart Labs

show abstract

Neural Graph Collaborative Filtering

Cited by 2,337 publications

References 30 publications

Learning to Hash with Graph Neural Networks for Recommender Systems

Learning to Hash with Graph Neural Networks for Recommender Systems

Reinforced Negative Sampling over Knowledge Graph for Recommendation

BasConv: Aggregating Heterogeneous Interactions for Basket Recommendation with Graph Convolutional Neural Network

Contact Info

Product

Resources

About