Online social networks (OSNs) are ubiquitous parts of modern societies, and millions of people use them on a daily basis to communicate with their friends and to stay up to date with recent news. However, recently OSNs have been accused of creating filter bubbles and increasing polarization and disagreement in societies. In this special interest group, we study these phenomena. Our goal is to understand how the timeline algorithms deployed in OSNs interact with the opinion formation processes in modern societies. Our ultimate aim is to develop effective regulatory measures for algorithms to minimize harm to individuals and societies.

From a more technical point of view, our work ranges from building abstract models of such phenomena to solving the involved computational challenges. In the process, we utilize tools from graph theory and graph mining. This allows us to simulate the impact of timeline algorithms on opinion formation models from sociology. This interdisciplinary approach will allow us to understand which properties benign algorithms for OSNs should have, addressing some of the most important research questions of the coming decade.