A Simple Post-Processing Step to Improve the Fairness of Collaborative Recommender Systems

Author(s): ___

Originally published on Towards AI.

Overview

In this article, I will describe an algorithm that can be applied as a post-processing step to alleviate the popularity bias inherent in collaborative filtering-based recommender systems. The content of this article is based on the work of [1]. The code to reproduce the results and figures can be found in this repository.

I assume the reader is familiar with the concept of collaborative filtering in the context of recommender systems.

Collaborative filtering-based recommender systems have a tendency to emphasize popular items i.e. items that receive many interactions from users (e.g. ratings, purchases, likes). This is problematic for the following reasons:

Not everyone… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI