Abstract
Recently, recommender systems based on Graph Convolution Network (GCN) have become a research hotspot, especially in collaborative filtering. However, most GCN-based models have inferior embedding propagation mechanism, leading to low information extraction efficiency. Besides, the existing methods suffer from high computational complexity for large user-item interaction graphs. In order to solve the above problems, we propose LII-GCCF that integrates Linear transformation, Initial residual and Identity mapping into the Graph Convolutional Collaborative Filtering model. First, initial residual and identity mapping are applied to optimize the information propagation of graph convolution, which privide abundant interaction and alleviate information loss problem. Second, LII-GCCF removes the unnecessary nonlinear transformation based on the characteristics of collaborative filtering to simplify the graph convolution process. Comprehensive experiments are conducted on two public datasets, and the results demonstrate that LII-GCCF has a significant improvement over other state-of-the-art methods.
Highlights
In the past few years, the rapid development of online services has led to a data explosion
To tackle the above problems, we propose LII-GCCF, which integrates Linear transformation, Initial residual and Identity mapping into the Graph Convolutional Collaborative Filtering model
3) BASELINES In order to verify the effectiveness of our model, we compare LII-GCCF with the following methods:
Summary
In the past few years, the rapid development of online services has led to a data explosion. Collaborative Filtering (CF) [5]–[8] is a fundamental algorithm for recommendation, which assumes that people with similar behaviors (e.g., watch the same video frequently) in the past have similar interests in items (e.g., music, videos, articles). A collaborative filtering model based on deep learning consists of two components: First, the embeddings of users and items are learned from the historical interactions. Based on the embeddings, the interaction function is used to generate personalized predictions for users. Traditional models such as matrix factorization (MF [9]) discard the connected relationship in the embedding function by projecting the ID of each user and item into a vector representation. Some follow-up methods consider part of the user-item interactions as features to enrich the embeddings of users or items, e.g., The associate editor coordinating the review of this manuscript and approving it for publication was Abdullah Iliyasu
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: 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.
