###
工程科学与技术:2024,56(1):44-53
←前一篇   |   后一篇→
本文二维码信息
码上扫一扫!
基于跨视图对比学习的知识感知推荐系统
(1.四川大学 计算机学院,四川 成都 610065;2.中国电子科技集团公司第三十研究所,四川 成都 610225)
Knowledge-aware Recommender System with Cross-views Contrastive Learning
(1.School of Computer Sci., Sichuan Univ., Chengdu 610065, China;2.No. 30 Inst. of CETC, Chengdu 610065, China)
摘要
图/表
参考文献
相似文献
附件
本文已被:浏览 196次   下载 129
投稿时间:2023-06-06    
中文摘要: 知识感知推荐(KGR)领域普遍存在监督信号稀疏问题。为了解决这个问题,对比学习方法被越来越广泛地应用于KGR。但是,过去基于对比学习的KGR模型仍存在一些问题:首先,使用图卷积对所有邻居节点直接聚合,无法排除知识图谱中不必要邻居节点信息的干扰;此外,只关注全局视图的信息,忽略了局部特征,这会导致过平滑问题。为了解决以上问题,提出一种基于跨视图对比学习的知识感知推荐系统(knowledge-aware recommender system with cross-views contrastive learning,KRSCCL)。KRSCCL使用关系图注意力网络构建包含用户、物品和实体节点的全局视图;使用轻量级图卷积网络构建包含用户和物品节点的局部视图,强调局部特征,有效地缓解过平滑问题;最后,在构建的两个视图的图内和图间节点对之间进行对比学习,以充分提取知识图谱信号,优化用户和物品表示。为了验证模型的有效性,在3个不同领域的公开数据集上进行了实验,实验结果表明:关系图注意力网络可以有效排除复杂网络聚合时的噪声问题;引入局部视图可以优化节点表示生成,缓解过平滑问题;KRSCCL模型在这3个数据集上都表现良好,在电影领域数据集Movielens–1M上,推荐的评估指标F1分数较最强基线提升2.0%;在音乐领域数据集Last.FM上,F1分数较最强基线提升0.3%;在书籍领域数据集Book–Crossing上,F1分数较最强基线提升5.1%。证明了本文模型的有效性。
Abstract:The knowledge-aware recommendation (KGR) domain commonly suffers from the problem of supervised signal sparsity, and contrast learning methods are increasingly studied to address this issue. However, existing contrast learning-based KGR models still have the following limitations. First, existing methods failed to suppress the interference information of unnecessary neighbouring nodes in the knowledge graph because graph convolution is used to directly aggregate all neighbouring nodes; Second, focusing only on the global information would lead to ignoring the fine-grained local features, causing over-smooth issues. In this work, a Knowledge-aware Recommender System with Cross-Views Contrastive Learning (KRSCCL) is proposed to address the aforementioned issues. In the KRSCCL, a relational graph attention network is proposed to construct a global view, including user, item and entity nodes. A lightweight graph convolutional network is designed to construct a local view, including user and item nodes, in which local features are emphasized to effectively mitigate the over-smooth problem. Finally, the contrastive learning mechanism is performed between intra- and inter-graph node pairs of the two views to fully extract KG signals and further optimize the user and item representations. Experimental results on three public datasets from different domains demonstrate that the proposed KRSCCL achieves expected performance improvement on all the three datasets over selective baselines, F1 score improvement on Movielens-1M, Last.FM and Book-crossing are 2.0%, 0.3% and 5.1%, respectively. Most importantly, the relational graph attention network can effectively exclude the noise during the feature aggregation of complex networks, the local views can optimize the generation of the node representation and alleviate the over-smooth problem.
文章编号:202300431     中图分类号:TP391    文献标志码:
基金项目:国家自然科学基金项目(62137001)
作者简介:第一作者:鄢凡力(1999-),男,硕士生.研究方向:推荐系统.E-mail:yanfanli@stu.scu.edu.cn;通信作者:琚生根,教授,E-mail:jsg@scu.edu.cn
引用文本:
鄢凡力,胥小波,赵容梅,孙思雨,琚生根.基于跨视图对比学习的知识感知推荐系统[J].工程科学与技术,2024,56(1):44-53.
YAN Fanli,XU Xiaobo,ZHAO Rongmei,SUN Siyu,JU Shenggen.Knowledge-aware Recommender System with Cross-views Contrastive Learning[J].Advanced Engineering Sciences,2024,56(1):44-53.