arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1901.08907cs.IRstat.ML
收藏

面向知识图谱增强推荐的多任务特征学习

Multi-Task Feature Learning for Knowledge Graph Enhanced Recommendation

Hongwei Wang, Fuzheng Zhang, Miao Zhao, Wenjie Li, Xing Xie, Minyi Guo

摘要:

协同过滤在实际的推荐场景中经常会遇到稀疏性和冷启动问题,因此研究人员和工程师通常使用边信息来解决这些问题,提高推荐系统的性能。在本文中,我们考虑知识图谱作为边信息的来源。我们提出了MKR,一个多任务的知识图谱增强推荐的特征学习方法。MKR是一个深度的端到端框架,它利用知识图谱嵌入任务来辅助推荐任务。这两个任务通过cross&compress单元相关联,cross & compress单元自动共享潜在特征并学习推荐系统中的项目与知识图谱中的实体之间的高阶交互。我们证明了交叉压缩单元具有足够的多项式逼近能力,并表明MKR是推荐系统和多任务学习的几个代表性方法的广义框架。通过对真实世界数据集的广泛实验,我们证明了MKR在电影,书籍,音乐和新闻推荐方面取得了巨大的进步,超过了最先进的基线。MKR也被证明能够保持一个体面的性能,即使用户项交互稀疏。

英文摘要:

Collaborative filtering often suffers from sparsity and cold start problems in real recommendation scenarios, therefore, researchers and engineers usually use side information to address the issues and improve the performance of recommender systems. In this paper, we consider knowledge graphs as the source of side information. We propose MKR, a Multi-task feature learning approach for Knowledge graph enhanced Recommendation. MKR is a deep end-to-end framework that utilizes knowledge graph embedding task to assist recommendation task. The two tasks are associated by cross&compress units, which automatically share latent features and learn high-order interactions between items in recommender systems and entities in the knowledge graph. We prove that cross&compress units have sufficient capability of polynomial approximation, and show that MKR is a generalized framework over several representative methods of recommender systems and multi-task learning. Through extensive experiments on real-world datasets, we demonstrate that MKR achieves substantial gains in movie, book, music, and news recommendation, over state-of-the-art baselines. MKR is also shown to be able to maintain a decent performance even if user-item interactions are sparse.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑