ConsisRec: Enhancing GNN for Social Recommendation via Consistent Neighbor Aggregation

Yang, Liangwei; Liu, Zhiwei; Dou, Yingtong; Ma, Jing; Yu, Philip S.

doi:10.1145/3404835.3463028

Cited by 107 publications

(35 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We review some graph-based recommendation methods as we adopt the graph neural network (GNN) [38,43,46,54] for the game recommendation. Graph-based recommendation methods model useritem interactions as a bipartite graph [2,14,23,32,51,53], with potential extensions to the heterogeneous graph with additional user-user social graph [25,49] and item knowledge graph [39,40]. Graph-based methods mostly adopt GNN for learning nodes (users and items) embeddings.…”

Section: Graph-based Recommendationmentioning

confidence: 99%

“…DANSER [45] performs dual graph attention networks on social and user-item interaction networks separately and fuses the learned embedding by a policy-based fusion layer. ConsisRec [49] dynamically samples informative neighbors and performs aggregation considering different relation types. FeSoG [21] proposes a GNN-based social recommendation system under graph federated learning setting [13].…”

Section: Graph-based Recommendationmentioning

confidence: 99%

“…Junliang et al [52] discuss that explicit social links are not all reliable due to the existence of spammers and bots. Also, the inconsistent social neighbors [49] can also be seen as noise. When we pay much importance weight to social embedding, the noise in social neighbors is also amplified, which may lower down the performance.…”

Section: Social and Context Influencementioning

confidence: 99%

“…In this sense, discovering potential interests of users from their friends enables the recommender system to identify the game engagements of users from a social perspective. Nevertheless, the challenge is that due to social inconsistency [49], users' friends may be of different impacts concerning distinct games. For example, users usually play MOBA games Dota 2 4 with friends in a team, whereas single-player RPG games Portal 5 are played individually.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Large-scale Personalized Video Game Recommendation via Social-aware Contextualized Graph Neural Network

Yang,

Liu,

Wang

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Because of the large number of online games available nowadays, online game recommender systems are necessary for users and online game platforms. The former can discover more potential online games of their interests, and the latter can attract users to dwell longer in the platform. This paper investigates the characteristics of user behaviors with respect to the online games on the Steam platform. Based on the observations, we argue that a satisfying recommender system for online games is able to characterize: personalization, game contextualization and social connection. However, simultaneously solving all is rather challenging for game recommendation. Firstly, personalization for game recommendation requires the incorporation of the dwelling time of engaged games, which are ignored in existing methods. Secondly, game contextualization should reflect the complex and high-order properties of those relations. Last but not least, it is problematic to use social connections directly for game recommendations due to the massive noise within social connections. To this end, we propose a Social-aware Contextualized Graph Neural Recommender System (SCGRec), which harnesses three perspectives to improve game recommendation. We conduct a comprehensive analysis of users' online game behaviors, which motivates the necessity of handling those three characteristics in the online game recommendation.

show abstract

Section: Graph-based Recommendationmentioning

confidence: 99%

Section: Graph-based Recommendationmentioning

confidence: 99%

Section: Social and Context Influencementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Large-scale Personalized Video Game Recommendation via Social-aware Contextualized Graph Neural Network

Yang,

Liu,

Wang

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The overload of information hinders users' ability to find what they really need among a large amount of items. To enhance users' experience, recommender systems have been applied in many web applications including e-commerce [1], [2], social recommendations [3], [4], and movie recommendations [5].…”

Section: Introductionmentioning

confidence: 99%

Pre-training Recommender Systems via Reinforced Attentive Multi-relational Graph Neural Network

Li¹,

Liu²,

Guo³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Recently, Graph Neural Networks (GNNs) have proven their effectiveness for recommender systems. Existing studies have applied GNNs to capture collaborative relations in the data. However, in real-world scenarios, the relations in a recommendation graph can be of various kinds. For example, two movies may be associated either by the same genre or by the same director/actor. If we use a single graph to elaborate all these relations, the graph can be too complex to process. To address this issue, we bring the idea of pre-training to process the complex graph step by step. Based on the idea of divideand-conquer, we separate the large graph into three sub-graphs: user graph, item graph, and user-item interaction graph. Then the user and item embeddings are pre-trained from user and item graphs, respectively. To conduct pre-training, we construct the multi-relational user graph and item graph, respectively, based on their attributes.In this paper, we propose a novel Reinforced Attentive Multirelational Graph Neural Network (RAM-GNN) to pre-train user and item embeddings on the user and item graph prior to the recommendation step. Specifically, we design a relation-level attention layer to learn the importance of different relations. Next, a Reinforced Neighbor Sampler (RNS) is applied to search the optimal filtering threshold for sampling top-k similar neighbors in the graph, which avoids the over-smoothing issue. We initialize the recommendation model with the pre-trained user/item embeddings. Finally, an aggregation-based GNN model is utilized to learn from the collaborative relations in the user-item interaction graph and provide recommendations. Our experiments demonstrate that RAM-GNN outperforms other state-of-the-art graph-based recommendation models and multirelational graph neural networks.

show abstract

Recommendation Model Based on Social Homogeneity Factor and Social Influence Factor

Ying

Wang

2021

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

ConsisRec: Enhancing GNN for Social Recommendation via Consistent Neighbor Aggregation

Cited by 107 publications

References 36 publications

Large-scale Personalized Video Game Recommendation via Social-aware Contextualized Graph Neural Network

Large-scale Personalized Video Game Recommendation via Social-aware Contextualized Graph Neural Network

Pre-training Recommender Systems via Reinforced Attentive Multi-relational Graph Neural Network

Recommendation Model Based on Social Homogeneity Factor and Social Influence Factor

Contact Info

Product

Resources

About