Evaluating the Use of User Content Feed Swapping for Counteracting Filter Bubbles

Abstract

The term “filter bubble” refers to a phenomenon in which a social media recommendation system fails to offer diverse or novel content, and instead offers content that reinforces particular belief systems. Filter bubbles are considered harmful because of their potential polarizing effects in society and their role in the spread of false information online. In this paper, we propose a solution to counteract the effects of filter bubbles by providing users with the option to switch content feeds with their least similar user’s feed. This is achieved by substituting the correlation coefficient used in collaborative filtering recommendation systems. A social media network simulation and accompanying questionnaire were used to test the viability of the solution. It was found to be viable in a simulated environment because it increased the users’ self-reported bias perception, without adversely impacting user engagement metrics, after switching with their least similar user’s feed. While a viable proof of concept in a simulated environment, the solution must be tested within a naturalistic setting with more participants in order to determine its real-world viability.