Abstract
In this work, we focus on dealing with a sparse users' feedback matrix and short descriptions/contents of items in recommender systems. We propose the Neural Poisson factorization (NPF) model which is a hybrid of deep learning and Poisson factorization. While Poisson factorization is suitable to model discrete, massive and sparse feedback, using a deep neural network and pre-trained word embeddings can learn hidden semantic in short item descriptions well. Therefore, NPF overcomes the limitation of existing models when dealing with short texts and a sparse feedback matrix. Moreover, we develop a random view algorithm based on stochastic learning for our model, in which each user is only viewed a random subset of items and his/her feedback on the subset is used to update his/her representation in each iteration. This approach is reasonable because each user can only know or view a partial subset of items when surfing a system. Extensive experiments illustrate the significant advantages of NPF over content-based matrix factorization methods and others that ignore item descriptions.
Highlights
The recommendation is an important component of a system when users increasingly demand convenient services
We focus on building a model that takes advantages of both deep neural networks for item contents and Poisson factorization for a feedback matrix
We propose a novel model, Neural Poisson factorization (NPF), that incorporates a neural network into Poisson factorization in order to deal with a sparse feedback matrix and short item contents
Summary
The recommendation is an important component of a system when users increasingly demand convenient services. The more the system develops and extends, the more users overload with data. People often explore several items to discover the desired ones. A recommender system necessarily detects and displays what users interest before they leave. Content-based matrix factorization, a hybrid of collaborative and content-based filtering, emerges as a promising method for personalized recommendation. These methods belonging to the group of content-based matrix factorization utilize matrix factorization as a core and enrich it by incorporating a model of item content (such as descriptions, reviews, synopses, abstracts, categories)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
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.