Transferable knowledge transfer for recommender systems

作者：禅与计算机程序设计艺术

1.背景介绍

Recommender Systems (RSs), also known as collaborative filtering RSs, are widely used in various applications such as e-commerce platforms, music streaming services, social networking sites etc., where the users’ preferences towards different items or products can be inferred from their past interactions with other users or system generated content. However, a significant challenge is to transfer the learned user preferences from one application context to another application context where they might be slightly different due to differences in environmental conditions, interests, culture, personalities, age groups, and many others. To address this issue, recent research has focused on learning transferable representations of user prefe