4.6 Article

Disentangled Item Representation for Recommender Systems

出版社

ASSOC COMPUTING MACHINERY
DOI: 10.1145/3445811

关键词

Representation learning; recommender systems; attribute disentangling

资金

  1. National Key Research and Development Program [2018YFB1402605, 2018YFB1402600]
  2. National Natural Science Foundation of China [U19B2038, 61772528]
  3. Beijing National Natural Science Foundation [4182066]

向作者/读者索取更多资源

Item representations in recommendation systems are traditionally done using single latent vectors, but utilizing attribute information has recently become popular for better item representations. This article proposes a fine-grained Disentangled Item Representation (DIR) method, representing items as separate attribute vectors for more detailed item information. Experimental results using the LearnDIR strategy show that models developed under DIR framework are effective and efficient, even outperforming state-of-the-art methods in cold-start situations.
Item representations in recommendation systems are expected to reveal the properties of items. Collaborative recommender methods usually represent an item as one single latent vector. Nowadays the e-commercial platforms provide various kinds of attribute information for items (e.g., category, price, and style of clothing). Utilizing this attribute information for better item representations is popular in recent years. Some studies use the given attribute information as side information, which is concatenated with the item latent vector to augment representations. However, the mixed item representations fail to fully exploit the rich attribute information or provide explanation in recommender systems. To this end, we propose a fine-grained Disentangled Item Representation (DIR) for recommender systems in this article, where the items are represented as several separated attribute vectors instead of a single latent vector. In this way, the items are represented at the attribute level, which can provide fine-grained information of items in recommendation. We introduce a learning strategy, LearnDIR, which can allocate the corresponding attribute vectors to items. We show how DIR can be applied to two typical models, Matrix Factorization (MF) and Recurrent Neural Network (RNN). Experimental results on two real-world datasets show that the models developed under the framework of DIR are effective and efficient. Even using fewer parameters, the proposed model can outperform the state-of-the-art methods, especially in the cold-start situation. In addition, we make visualizations to show that our proposition can provide explanation for users in real-world applications.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.6
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据