Discovery and classification of Twitter bots

Abstract

A very large number of people use Online Social Networks daily. Such platforms thus become attractive targets for agents that seek to gain access to the attention of large audiences, and influence perceptions or opinions. Botnets, collections of automated accounts controlled by a single agent, are a common mechanism for exerting maximum influence. Botnets may be used to better infiltrate the social graph over time and to create an illusion of community behavior, amplifying their message and increasing persuasion. This paper investigates Twitter botnets, their behavior, their interaction with user communities and their evolution over time. We analyzed a dense crawl of a subset of Twitter traffic, amounting to nearly all interactions by Greek-speaking Twitter users for a period of 36 months. We detected over a million events where seemingly unrelated accounts tweeted nearly identical content at nearly the same time. We filtered these concurrent content injection events and detected a set of 1,850 accounts that repeatedly exhibit this pattern of behavior, suggesting that they are fully or in part controlled and orchestrated by the same software. We found botnets that appear for brief intervals and disappear, as well as botnets that evolve and grow, spanning the duration of our dataset. We analyze statistical differences between bot accounts and human users, as well as botnet interaction with user communities and Twitter trending topics.