Measuring the Impact of Recommendation Algorithms on Misinformation Diffusion in Social Networks

Misinformation in social media is increasing drastically through various factors including bots, adaptive user behavior, content constructions, and the recommendation algorithms that social media platforms use. While recent work focuses on understanding misinformation from user content and bots’ predictions, works on the effect of recommendation algorithms to amplify misinformation have not received much attention. In this talk, we present an in-depth study which aims to understand the interconnection between misinformation spread and recommendation algorithms. We conducted two different experiments using two datasets. One to understand the performance of recommendation algorithms in the presence of misinformation and explore which algorithm is more prone to suggest misinformation. The second one to investigate the role of recommendation algorithms on misinformation propagation in social networks. We will discuss emerging trends and offer insights on the impact of misinformation on different recommendation algorithms that researchers and practitioners can adopt.