arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.04797cs.AI
收藏

Attentive Social Recommendation:面向用户和项目的配送

Attentive Social Recommendation: Towards User And Item Diversities

The Pennsylvania State University(宾夕法尼亚州立大学);Baidu Research(百度研究院);StylingAI Inc.(StylingAI公司)

Dongsheng Luo, Yuchen Bian, Xiang Zhang, Jun Huan

摘要:

社会推荐系统是利用用户与用户之间的社会关系和用户与项目之间的评分来预测用户与项目之间的评分值。然而,在社会推荐中的用户/项目差异在文献中没有得到很好的利用。特别是,因素间(社会和评级因素)的关系和独特的评级价值需要更多的考虑。在本文中,我们提出了一个关注的社会推荐系统(ASR)来解决这个问题,从两个方面。首先,在ASR中,Rec-conv图网络层提出提取社会因素,用户评级和项目评级因素,然后自动分配贡献权重,将这些因素聚合到用户/项目嵌入向量。第二,对不同的评级值应用了一种分解策略。大量的基准测试实验证明了我们的ASR的有效性和优势。

英文摘要:

Social recommendation system is to predict unobserved user-item rating values by taking advantage of user-user social relation and user-item ratings. However, user/item diversities in social recommendations are not well utilized in the literature. Especially, inter-factor (social and rating factors) relations and distinct rating values need taking into more consideration. In this paper, we propose an attentive social recommendation system (ASR) to address this issue from two aspects. First, in ASR, Rec-conv graph network layers are proposed to extract the social factor, user-rating and item-rated factors and then automatically assign contribution weights to aggregate these factors into the user/item embedding vectors. Second, a disentangling strategy is applied for diverse rating values. Extensive experiments on benchmarks demonstrate the effectiveness and advantages of our ASR.

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。

↑