User Recommendation in Content Curation Platforms
User Recommendation in Content Curation Platforms
复制标题
DOI：
10.1145/3336191.3371822
复制
发表时间：
2020-01
期刊：
影响因子：
--
通讯作者：
Jianling Wang;Ziwei Zhu;James Caverlee
中科院分区：
文献类型：
--
作者：
Jianling Wang;Ziwei Zhu;James Caverlee
我们为内容策展平台提出了一个个性化用户推荐框架，该框架同时对用户及其所参与项目的偏好进行建模。通过这种方式，用户对特定项目类型（例如奇幻小说）的偏好可以与用户的专长（例如评论以强大女性为主角的小说）相平衡。特别地，所提出的模型具有三个独特特征：（i）它通过一个多方面自动编码器模型同时学习用户 - 项目和用户 - 用户偏好；（ii）它融合用户在用户和项目上偏好的潜在表示，通过一个对抗框架构建共享因素；（iii）它包含一个注意力层，以产生不同潜在表示的加权聚合，从而改进用户和项目的个性化推荐。通过与最先进的模型进行实验，我们发现所提出的框架在 top - k用户推荐方面分别在Goodreads上提高了18.43%，在Spotify上提高了6.14%。
We propose a personalized user recommendation framework for content curation platforms that models preferences for both users and the items they engage with simultaneously. In this way, user preferences for specific item types (e.g., fantasy novels) can be balanced with user specialties (e.g., reviewing novels with strong female protagonists). In particular, the proposed model has three unique characteristics: (i) it simultaneously learns both user-item and user-user preferences through a multi-aspect autoencoder model; (ii) it fuses the latent representations of user preferences on users and items to construct shared factors through an adversarial framework; and (iii) it incorporates an attention layer to produce weighted aggregations of different latent representations, leading to improved personalized recommendation of users and items. Through experiments against state-of-the-art models, we find the proposed framework leads to a 18.43% (Goodreads) and 6.14% (Spotify) improvement in top-k user recommendation.