A deeper graph neural network for recommender systems☆
Introduction
Recommender systems have become increasingly important in recent years due to the problem of information overload. Recommender systems allow individuals to acquire information more effectively by filtering information. Over the years, collaborative filtering has become the most successful and widely used recommendation technique [1], [2]. The core assumption is that users who have expressed similar interests in the past will share similar interests in the future. Popularized by the Netflix Prize, Matrix Factorization (MF) has become the de facto approach to collaborative filtering-based recommendation. Much research effort has been devoted to enhancing the MF method, one of the most powerful collaborative filtering techniques, such as integrating it with neighbor-based models [3], extending it to factorization machines for the generic modeling of features [4], and optimizing it with Bayesian personalized ranking objective to adapt it to implicit feedback recommendation [5]. However, despite these efforts, sparsity is still one of the most challenging issues facing us today.
Recent years have seen a surge in research on graph neural networks, leading to substantial improvements in the performance of tasks with graph-structured data, which is fundamental for recommendation applications. One of the most prominent approaches is Graph Convolutional Networks (GCNs) [6]. The core idea behind GCNs is finding a way to iteratively aggregate feature information about graph structure and the structure of the node’s local graph neighborhood into a machine learning model. The goal is to learn a mapping that embeds nodes as points in a low-dimensional vector space . The primary contribution of representation learning approaches is that of finding a way to represent, or encode graph structure which geometric relationships in the embedding space reflect the structure of the original graph. However, GCNs need to operate on the Laplacian eigenbasis which leads to a huge time consumption on large graphs .
In this paper, we view the recommendation task as a link prediction problem on a bipartite graph: the interaction data in collaborative filtering-based methods can be represented by a bipartite graph between user and item nodes as shown in 1. The MF approaches can then be considered as learning a mapping from users/items to a low-dimensional vector, where the interaction information is contained in the vectors.
We propose a general framework named GCF, short for Graph neural network-based Collaborative Filtering, which builds on recent progress in graph neural networks. The framework contains a larger receptive field with iterative information propagation enabling our method to access more information in the process of making decisions. The concept of receptive field on graph neural networks will be given in the following section.
We present an attention-based message-passing method to carry out the information propagation process. In the recommendation task, the variable size input for each layer is a challenge, because the number of neighbors for each node is different. To solve this problem, we assign different weights for neighbors.
Our contributions are as follows:
- •
We present a general framework GCF to model the latent features of users and items. We also show that MF is a special case of GCF when the number of hidden layers is no more than one.
- •
We propose an attention-based message-passing method to solve the variable size input problem.
- •
We perform extensive experiments on three real-world datasets to demonstrate the performance of our GCF method. The results show that the proposed method outperforms the state-of-the-art methods in terms of HR@k and NDCG@k.
The rest of this paper is organized as follows. In Section 2, we formalize the problem and review some of the relevant methods. Section 3 presents our approach in detail. The experiments and results analysis are demonstrated in Section 4. Lastly, conclusions and future work are discussed in Section 5.
Section snippets
Preliminaries and literature review
In this section, We formalize the problem and discuss existing solutions for collaborative filtering with implicit feedback.
Graph neural network-based collaborative filtering
In this section, we first present the general GCF framework. We then show that SVD and SVD can be expressed under GCF with node embedding via graph neural network. To address the problem of dealing with variable size inputs in the information propagation process, we propose a new method with an attention mechanism which assigns different weights to the neighbors of each node.
Experimental settings
Experiments are conducted on three datasets namely MovieLens 1M, MovieLens 10M and Taobao. The basic statistics are listed in Table 1. We select 60% of records as the training set. Some records contain explicit feedback such as ratings. As our focus is on the implicit feedback task, we remove the ratings from these datasets.
MovieLens is a common benchmark dataset which consists of user ratings for items. Many versions have been released on the GroupLens website. We select MovieLens 1M (ML-1M)
Conclusions and further study
In this study, we have introduced a general framework GCF and an information propagation-based graph neural network. GCF is a representation learning framework for learning a mapping that embeds users and items as points in a low-dimensional vector space with geometric relationships in the embedding space that reflect the preference relationship between users and items. To address the problem of variable size inputs for each node on a bipartite graph, we have proposed an attention-based
Acknowledgments
This research was supported by National Key R&D Program of China (No. 2016YFB0801100), Beijing Natural Science Foundation, China (No. 4172054, No. L181010), and National Basic Research Program of China (No. 2013CB329605). This work was also supported by the Australian Research Council (ARC) under Grant [DP170101632].
References (25)
- et al.
Member contribution-based group recommender system
Decis. Support Syst.
(2016) - et al.
Recommender system application developments: A survey
Decis. Support Syst.
(2015) - et al.
Recommender systems survey
Knowl.-Based Syst.
(2013) - et al.
A cross-domain recommender system with consistent information transfer
Decis. Support Syst.
(2017) - et al.
Collaborative filtering beyond the user-item matrix: A survey of the state of the art and future challenges
ACM Comput. Surv.
(2014) - et al.
A fuzzy preference tree-based recommender system for personalized business-to-business e-services
IEEE Trans. Fuzzy Syst.
(2015) Factorization meets the neighborhood: a multifaceted collaborative filtering model
Factorization machines
- et al.
BPR: Bayesian personalized ranking from implicit feedback
- et al.
Convolutional neural networks on graphs with fast localized spectral filtering
Multirelational social recommendations via multigraph ranking
IEEE Trans. Cybern.
The bellkor solution to the netflix grand prize
Netflix Prize Doc.
Cited by (112)
Migrate demographic group for fair Graph Neural Networks
2024, Neural NetworksPersonalized Image Aesthetics Assessment based on Graph Neural Network and Collaborative Filtering
2024, Knowledge-Based SystemsAn interpretable approach using hybrid graph networks and explainable AI for intelligent diagnosis recommendations in chronic disease care
2024, Biomedical Signal Processing and ControlCognitive-based knowledge learning framework for recommendation
2024, Knowledge-Based SystemsJoint item recommendation and trust prediction with graph neural networks
2024, Knowledge-Based SystemsNeuro-symbolic recommendation model based on logic query
2024, Knowledge-Based Systems
- ☆
No author associated with this paper has disclosed any potential or pertinent conflicts which may be perceived to have impending conflict with this work. For full disclosure statements refer to https://doi.org/10.1016/j.knosys.2019.105020.