The paper "A Proposal for Uncovering Hidden Social Bots via Genetic Similarity" has been accepted as a poster at Discovery Science 2024!

Brief summary

In this short paper, we propose a novel method to address the growing challenge of social bots on social media platforms. Social bots, often responsible for spreading disinformation and manipulating public opinion, have become increasingly sophisticated, with some now utilizing large language models to mimic human behavior. Our approach introduces the use of biological similarity algorithms to classify social media users as bots or genuine accounts. By analyzing user behavior, we aim to cluster users into macro species based on timeline similarities, which are then further classified using genetic characteristics. This adaptive method allows detection criteria to evolve alongside the ever-changing tactics of bots. Additionally, the idea we present will guide towards existing detection applications based on timeline equality alone to be extended to detect bots. By incorporating new metrics, our approach will systematically classify non-trivial accounts into appropriate categories, effectively peeling back layers to reveal non-obvious species.

Stay tuned for more updates as this research progresses!