Unifying Task-Oriented Knowledge Graph Learning and Recommendation
Unifying Task-Oriented Knowledge Graph Learning and Recommendation
复制标题
统一面向任务的知识图谱学习和推荐
DOI：
10.1109/access.2019.2932466
复制
发表时间：
2019
期刊：
影响因子：
3.9
通讯作者：
Song Hengjie
中科院分区：
文献类型：
--
作者：
Li Qianyu;Tang Xiaoli;Wang Tengyun;Yang Haizhi;Song Hengjie
将知识图谱（knowledge graph, KGs）整合到推荐系统（知识感知推荐）中以提高推荐的准确性和可解释性已经引起了大量的研究。然而，现有的方法在很大程度上假设KGs在转移知识时是完整的，这可能导致这些KGs的性能不理想，在现实场景中几乎是不完整的。在本文中，我们提出了一个鲁棒共同学习模型（RCoLM），该模型在将KGs纳入推荐时考虑了它们的不完全性。RCoLM的目的是利用迁移学习模型在推荐任务和知识图完成任务之间进行知识迁移。本文的早期版本出现在KDD 2019上。这个版本是先前提交的扩展，这里提出了两个主要的创新。首先，与以往的知识感知推荐方法不同，RCoLM方法主要侧重于将知识从KG转移到项目推荐，而RCoLM方法试图利用推荐中的用户-项目交互来完成KG，并将这两个任务统一在一个联合模型中进行相互增强。其次，RCoLM在KG补全任务上提供了一种通用的面向任务的负抽样策略，进一步提高了算法的自适应能力，对于在KG补全的各个子任务中获得优异的性能起着至关重要的作用。在两个真实的公共数据集上进行的大量实验表明，RCoLM不仅优于最先进的知识感知推荐方法，而且优于现有的KG补全方法。
Incorporating knowledge graphs (KGs) into recommender systems (knowledge-aware recommendation) to improve the recommendation accuracy and explainability has attracted considerable research efforts. However, existing methods largely assume that KGs are complete when transferring knowledge from them, which may lead to suboptimal performance for those KGs, can be hardly complete in real-life scenarios. In this paper, we present a robustly co-learning model (RCoLM) that takes the incompleteness nature of KGs into consideration when incorporating them into recommendation. The RCoLM aims at transferring knowledge between recommendation task and knowledge graph completion (KG completion) task by utilizing a transfer learning model. An earlier version of this paper appeared in KDD 2019. This version is an extension of the previous submission and two major innovations are presented here. At first, distinct from previous knowledge-aware recommendation methods, which mainly focus on transferring knowledge from KGs to item recommendations, the RCoLM attempts to exploit user-item interactions from recommendations for KG completion, and unifies the two tasks in a joint model for mutual enhancements. Second, the RCoLM provides a general task-oriented negative sampling strategy on KG completion task, which further improves the adaptive ability of the proposed algorithm and plays an essential role for obtaining superior performance in various sub-tasks of the KG completion. The extensive experiments on two real-world public datasets demonstrate that RCoLM outperforms not only state-of-the-art knowledge-aware recommendation methods but also existing KG completion methods.
登录
查看更多内容
DOI：
10.1109/wi-iat.2012.26
发表时间：
2012-12
期刊：
2012 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology
影响因子：
--
作者：
Zhicheng Zheng;Xiance Si;Fangtao Li;E. Chang;Xiaoyan Zhu
通讯作者：
Zhicheng Zheng;Xiance Si;Fangtao Li;E. Chang;Xiaoyan Zhu
DOI：
10.1145/2245276.2245341
发表时间：
2012-03
期刊：
--
影响因子：
--
作者：
Lucas Drumond;Steffen Rendle;L. Schmidt-Thieme
通讯作者：
Lucas Drumond;Steffen Rendle;L. Schmidt-Thieme
DOI：
--
发表时间：
2016-03
期刊：
ArXiv
影响因子：
--
作者：
QUAN LIU;Hui Jiang;Zhenhua Ling;Si Wei;Yu Hu
通讯作者：
QUAN LIU;Hui Jiang;Zhenhua Ling;Si Wei;Yu Hu
DOI：
10.1145/2339530.2339611
发表时间：
2012-08
期刊：
--
影响因子：
--
作者：
Ke Zhou;H. Zha
通讯作者：
Ke Zhou;H. Zha
DOI：
10.1145/2187836.2187874
发表时间：
2012-04
期刊：
Proceedings of the 21st international conference on World Wide Web
影响因子：
--
作者：
Maximilian Nickel;Volker Tresp;H. Kriegel
通讯作者：
Maximilian Nickel;Volker Tresp;H. Kriegel