Diversity exposure in recommender systems to approach filter bubbles and unfairness

• Grant number: 22/07016-9
• Support Opportunities: Regular Research Grants
• Start date: September 01, 2022
• End date: August 31, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Marcelo Garcia Manzato
• Grantee: Marcelo Garcia Manzato
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

Recommender systems have been researched extensively over the past decades to filter information. Current research is often focused on incremental improvements of algorithms, but overlooks critical challenges, including filter bubbles and unfairness caused by biased recommendations. Although there are efforts to reduce each of these issues individually, there is a lack of works that explore both problems at the same time. The overall goal of this project is to research how such filter bubbles and unfairness can be approached and solved. By efficiently combining both calibration and digital nudges towards more diverse recommendations, we plan to support awareness of filter bubbles and unfairness, and exposure to more diverse and fair content. (AU)