What is it about?
In practice, there are many related domains that may have data sparsity problems, but they can help each other to get great recommendation performance for themselves.
Featured Image
Why is it important?
In practice applications, data sparsity is widespread and damage recommendation performance.
Perspectives
This research idea comes from my personal experience. I observed that when we open a seldom-used application, the system usually recommends items that are similar to the ones we have interacted with before. Due to the limited observable data, the user preferences captured by the system are relatively one-sided. However, our preferences are generally multifaceted, and how to capture the user’s comprehensive preferences in this situation has become a research point I want to explore.
Zhao XIaoyun
Sichuan University
Read the Original
This page is a summary of: Multi-Sparse-Domain Collaborative Recommendation via Enhanced Comprehensive Aspect Preference Learning, February 2022, ACM (Association for Computing Machinery),
DOI: 10.1145/3488560.3498381.
You can read the full text:
Contributors
The following have contributed to this page
