Unravelling Filter Bubbles in Recommender Systems: A Comprehensive Review

Abstract

The prevalence of filter bubbles in recommender systems has raised concerns about the potential impact on user experiences and information exposure. This systematic literature review aims to deliver a thorough analysis for the study conducted on filter bubbles in recommender systems and its implications for user experiences and information exposure. Through a thorough bibliometric analysis using VOSViewer, the research landscape is mapped, influential authors are identified, and recurring themes are examined. The review delves into theoretical frameworks, empirical studies, and algorithmic approaches to highlight the challenges and opportunities associated with detecting and mitigating filter bubbles. Key findings encompass the identification of filter bubbles, their impact on user behavior, and proposed solutions. Additionally, an analysis of datasets used in this context is presented, alongside notable solutions developed to address filter bubbles in recommender systems. The review investigates the unresolved obstacles in the domain of filter bubbles within recommender systems. The review also explores potential future directions and solutions to mitigate filter bubbles, underscoring the significance of understanding their implications for user decision-making and information diversity. By contributing to the ongoing discourse on recommendation systems, this research offers valuable insights to researchers, practitioners, and policymakers seeking to address the filter bubble phenomenon and enhance user experiences in the realm of recommender systems.