Mitigating Filter Bubbles within Deep Recommender Systems
Machine Learning
2022-09-20 v1 Information Retrieval
Abstract
Recommender systems, which offer personalized suggestions to users, power many of today's social media, e-commerce and entertainment. However, these systems have been known to intellectually isolate users from a variety of perspectives, or cause filter bubbles. In our work, we characterize and mitigate this filter bubble effect. We do so by classifying various datapoints based on their user-item interaction history and calculating the influences of the classified categories on each other using the well known TracIn method. Finally, we mitigate this filter bubble effect without compromising accuracy by carefully retraining our recommender system.
Keywords
Cite
@article{arxiv.2209.08180,
title = {Mitigating Filter Bubbles within Deep Recommender Systems},
author = {Vivek Anand and Matthew Yang and Zhanzhan Zhao},
journal= {arXiv preprint arXiv:2209.08180},
year = {2022}
}
Comments
6 Pages 4 figures