优点:《Graph convolutional neural networks for web-scale recommender systems》,2018,被引3132,表明基于GNN的模型在实际平台上可以高效且有效地应用于大规模的推荐系统,也算是解决了缺点 其他 最近的一些研究移除了非线性关系,《LightGCN: Simplifying and powering graph convolution network for recommendation》,2020 ...
本文主要讨论了社交推荐(social recommendation)、序列推荐(seuqential recommendation)、会话推荐(session recommendation)、捆绑推荐(buddle recommendation)、跨域推荐(cross-domain recommendation)、多行为推荐(multi-behavior recommendation)这六种推荐场景。这里只对概念进行简要说明。 社交推荐(social recommendation) 在具有...
捆绑推荐,图源自论文《Graph Neural Networks for Recommender Systems: Challenges, Methods, and Directions》 该场景的主要论文: (5)跨域推荐 随着越来越多的用户跨域与多模式信息交互,跨域推荐(cross-domain recommendation, CDR)被证明是一种很有前景的解决冷启动和数据稀疏问题的方法。 跨域推荐,图源自论文《Graph...
跨域推荐(cross-domain recommendation) 跨域推荐会充分利用域和域之间的相关关系,可以缓解冷启动与数据稀疏问题。根据信息在域之间的流动方向,又可以把跨域推荐进一步分为从源域到目标域的single-target CDR(STCDR), 源域与目标域相互影响的dual-target CDR(DTCDR),和DTCDR的推广multi-target CDR(MTCDR)。 多行为...
Our experimental results demonstrate that our proposed method outperforms five baselines, suggesting that the interactions extracted help to alleviate the data sparsity problem and improve recommendation accuracy.doi:10.1016/j.knosys.2019.105020Ruiping Yin...
模型:Knowledge-aware Coupled Graph Neural Network (KCGN),在推荐框架中加入用户和项目相互依赖知识。利用全局图结构感知的相互信息,编码高阶用户和项目联系。 introduction 将社交联系加入协同过滤体系结构作为辅助信息,描述用户的连接性信息。 现存方法忽视了项目级别的信息,item-wise relational structure 有助于理解用...
Heterogeneous Graph Neural Network for Personalized Session-Based Recommendation with User-Sessio... 用户会话约束的个性化会话推荐异构图神经网络 该推荐系统为用户提供了一个适当限制的最近在线大量信息。最近,在这些会话中包含用户信息的研究正在取得进展。然而,很难生成包含用户生成的会话表示的高质量用户表示。在...
Graph Neural Networks for Social Recommendation 1. 摘要 构建基于图神经网络的推荐系统的三大挑战 the user-item graph encodes both interactions and their associated opinions social relations have heterogeneous strengths users involve in two graphs (e.g., the user-user social graph and the user-item ...
Graph Neural Networks for Social Recommendation 1. 摘要 构建基于图神经网络的推荐系统的三大挑战 the user-item graph encodes both interactions and their associated opinions social relations have heterogeneous strengths users involve in two graphs (e.g., the user-user social graph and the user-item ...
Graph neural networks (GNNs) have shown great potential for personalized recommendation. At the core is to reorganize interaction data as a user-item bipartite graph and exploit high-order connectivity among user and item nodes to enrich their representations. While achieving great success, most exist...