What is it about?
This paper introduces a new graph neural network (SKIPHOP) that makes recommender systems fairer by ensuring similar users get similar recommendations. It does this by looking at both direct user-item interactions and deeper connections in knowledge graphs and adding fairness constraints to the model. Experiments on real datasets show that SKIPHOP improves both fairness and accuracy.
Featured Image
Photo by Alexander Shatov on Unsplash
Why is it important?
Recommender systems shape what people see online, and unfair treatment, even if unintentional, can create unequal access to opportunities, information, or products. This work offers a method to balance fairness and accuracy, which could help make AI-driven recommendations more equitable for all users.
Read the Original
This page is a summary of: Equipping Recommender Systems with Individual Fairness via Second-order Proximity Embedding, November 2022, Institute of Electrical & Electronics Engineers (IEEE),
DOI: 10.1109/asonam55673.2022.10068703.
You can read the full text:
Contributors
The following have contributed to this page
