Abstract
Attribute-aware CF models aim at rating prediction given not only the historical rating given by users to items but also the information associated with users (e.g., age), items (e.g., price), and ratings (e.g., rating time). This paper surveys work in the past decade to develop attribute-aware CF systems and finds that they can be classified into four different categories mathematically. We provide readers not only with a high-level mathematical interpretation of the existing work in this area but also with mathematical insight into each category of models. Finally, we provide in-depth experiment results comparing the effectiveness of the major models in each category.
Highlights
Collaborative filtering is arguably the most effective method for building a recommender system
Model We consider several popular models that are representative of each category for comparison: the Regression-based Latent Factor Model (RLFM) (Agarwal and Chen, 2009) and FriendshipInterest Propagation (FIP) (Yang et al, 2011) in Discriminative Matrix Factorization, Collective Matrix Factorization (CMF) (Singh and Gordon, 2008) in Generative Matrix Factorization, and Tensor Factorization (TF) (Karatzoglou et al, 2010), the Factorization Machine (FM) (Rendle et al, 2011), and the Neural Factorization Machine (NFM) (He and Chua, 2017) in Generalized Factorization
FM can be seen as a design that incorporates attributes into matrix factorization (MF), and the same applies for Neural Collaborative Filtering (NCF)+ vs. NCF
Summary
Collaborative filtering is arguably the most effective method for building a recommender system. Researchers have proposed different methods to extend existing collaborative filtering models in recent years, such as factorization machines, probabilistic graphical models, kernel tricks, and models based on deep neural networks. We notice that those papers can be categorized based on the type of attributes incorporated into the models. Discriminative matrix factorization models extend the traditional MF by treating the attributes as prior knowledge to learn the latent representation of users or items. Generalized factorization models view the user/item identity as a kind of attribute, and various models have been designed for determining the lowdimensional representation vectors for rating prediction.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Similar Papers
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.