2 Gao et al.
the typical applications of machine learning driven by the real world, it is an extremely hot topic in
both industrial and academia nowadays.
To recap the history of recommender systems, it can be generally divided into three stages,
shallow models [
74
,
125
,
126
], neural models [
26
,
48
,
56
], and GNN-based models [
55
,
153
,
188
]. The
earliest recommendation models capture the collaborative ltering (CF) eect by directly calculating
the similarity of interactions. Then model-based CF methods, such as matrix factorization (MF) [
74
]
or factorization machine[
125
], were proposed to approach recommendation as a representation
learning problem. However, these methods are faced with critical challenges such as complex user
behaviors or data input. To address it, neural network-based models [
26
,
48
,
56
] are proposed. For
example, neural collaborative ltering (NCF) was developed to extend the inner product in MF
with multi-layer perceptrons (MLP) to improve its capacity. Similarly, deep factorization machine
(DeepFM) [
48
] combined the shallow model factorization machine (FM) [
125
] with MLP. However,
these methods are still highly limited since their paradigms of prediction and training ignore the
high-order structural information in observed data. For example, the optimization goal of NCF
is to predict user-item interaction, and the training samples include observed positive user-item
interactions and unobserved negative user-item interactions. It means that during the parameter
updating for a specic user, only the items interacted by him/her are involved.
Recently, the advances of graph neural networks provide a strong fundamental and opportunity
to address the above issues in recommender systems. Specically, graph neural networks adopt
embedding propagation to aggregate neighborhood embedding iteratively. By stacking the propaga-
tion layers, each node can access high-order neighbors’ information, rather than only the rst-order
neighbors’ as the traditional methods do. With its advantages to handle the structural data and
to explore structural information, GNN-based methods have become the new state-of-the-art
approaches in recommender systems.
To well apply graph neural networks into recommender systems, there are some critical chal-
lenges required to be addressed. First, the data input of recommender system should be carefully
and properly constructed to graph, with nodes representing elements and edges representing the
relations. Second, for the specic task, the component in the graph neural network should be
adaptively designed, including how to propagate and aggregate, in which existing works have
explored various choices with dierent advantages and disadvantages. Third, the optimization of
the GNN-based model, including the optimization goal, loss function, data sampling, etc., should
be consistent with the task requirement. Last, since recommender systems have strict limitations
on the computation cost, and also due to GNNs’ embedding propagation operations introduce
a number of computations, the ecient deployment of graph neural networks in recommender
systems is another critical challenge.
In this paper, we aim to provide a systematic and comprehensive review of the research eort,
especially on how they improve recommendation with graph neural networks and address the
corresponding challenges. To fulll a clear understanding, we categorize researches of recommender
systems from four perspectives, stage, scenario, objectives, and applications. We summarize the
representative papers along with their codes repositories in https://github.com/tsinghua-b-lab/
GNN-Recommender-Systems.
It is worth mentioning that there is one existing survey [
168
] of graph neural network-based
recommender system. However, it is limited due to the following reasons. First, it does not provide
extensive taxonomy of recommender systems. Specically, it roughly divides recommender systems
into non-sequential recommendation and sequential recommendation, however, which is not so
reasonable. In fact, the sequential recommendation is only one specic recommendation scenario
with a special setting of input and output, as pointed out by this survey. Second, it does not provide
adequate explanations of the motivations and reasons that the existing works leverage graph
ACM Transactions on Information Systems, Vol. 1, No. 1, Article . Publication date: September 2021.