What is it about?
Knowledge graphs, as side information, can effectively improve the performance of recommendation systems. However, not all relationships within the knowledge graph are necessarily relevant to downstream recommendation tasks. Irrelevant relationships may have a negative impact on the performance of the recommendation system. Therefore, we propose using a diffusion model to denoise the knowledge graph by filtering out irrelevant relationships, thereby enhancing the performance of the recommendation system.
Featured Image
Photo by CardMapr.nl on Unsplash
Why is it important?
We propose a novel diffusion denoising model on the knowledge graph, which can be guided by downstream recommendation tasks to filter out irrelevant relationships in the knowledge graph. This model aims to enhance the performance of knowledge-aware recommenders.
Perspectives
We hope that this work can contribute to the application of diffusion models in the field of recommendation systems. Currently, there are still many unexplored directions for the application of diffusion models in recommendation systems and graph-related domains.
Yangqin Jiang
University of Hong Kong
Read the Original
This page is a summary of: DiffKG: Knowledge Graph Diffusion Model for Recommendation, March 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3616855.3635850.
You can read the full text:
Resources
Contributors
The following have contributed to this page
