Online polarization has gained significant attention from researchers over the years, raising concerns about its societal effects. The design of personalized depolarization strategies is seen as a key solution, which must rely on precise measurements and a clear understanding of polarization behaviors. However, existing literature lacks fine-grained characterizations of these behaviors. We propose GRAIL, the first individual polarization metric based on multiple factors. GRAIL assesses these factors through entropy and is constructed on an adaptable Generalized Additive Model. We evaluate the proposed metric on a Twitter dataset related to the highly controversial debate about the COVID-19 vaccine, confirming GRAIL's ability to discriminate between polarization behaviors. Furthermore, we provide a finer characterization and explanation of the identified behaviors through an innovative evaluation framework.
Blogger's Review: The introduction of GRAIL offers a fresh perspective on analyzing polarization behaviors in social media. By integrating entropy with a generalized additive model, it captures user polarization features comprehensively, holding significant theoretical and practical implications. Future research could explore the application of this metric across different social media platforms.