There is a significant body of literature concerning the analysis of Twitter accounts, yet the behavior of newly created accounts remains relatively unexplored. In this study, we introduce a novel approach to detect Twitter accounts right after registration and explore their behavioral patterns. In a two-week period in April 2020, our technique identified over 500,000 accounts before they even started interacting with the platform. Each account was monitored for 21 days by sampling profile information and timelines at scheduled intervals, retrieving over 8 million tweets. An additional sample of profile information was collected approximately two years after creation, in May 2022. One of the key findings of our study is the lack of sustained and genuine engagement from new accounts. Indeed, a large proportion of them (almost 25%) were suspended by Twitter in the first 21 days, and the evaluation conducted after two years reveals that only a tiny fraction of the remaining enabled accounts seem to be active and genuine users (3.8% of the initial sample). Additionally, despite the early suspensions enforced by Twitter, it turns out that some short-lived accounts still managed to have a substantial impact on the total volume of content and interactions from new accounts. Overall, our findings may have important implications for understanding the dynamics of new accounts' behavior as well as Twitter's suspension policy prior to the recent change in ownership. This could stimulate further research to evaluate the impact of the ongoing changes introduced by the new administration.
Twitter Newcomers: Uncovering the Behavior and Fate of New Accounts Through Early Detection and Monitoring
G Cola;M Mazza;M Tesconi
2023
Abstract
There is a significant body of literature concerning the analysis of Twitter accounts, yet the behavior of newly created accounts remains relatively unexplored. In this study, we introduce a novel approach to detect Twitter accounts right after registration and explore their behavioral patterns. In a two-week period in April 2020, our technique identified over 500,000 accounts before they even started interacting with the platform. Each account was monitored for 21 days by sampling profile information and timelines at scheduled intervals, retrieving over 8 million tweets. An additional sample of profile information was collected approximately two years after creation, in May 2022. One of the key findings of our study is the lack of sustained and genuine engagement from new accounts. Indeed, a large proportion of them (almost 25%) were suspended by Twitter in the first 21 days, and the evaluation conducted after two years reveals that only a tiny fraction of the remaining enabled accounts seem to be active and genuine users (3.8% of the initial sample). Additionally, despite the early suspensions enforced by Twitter, it turns out that some short-lived accounts still managed to have a substantial impact on the total volume of content and interactions from new accounts. Overall, our findings may have important implications for understanding the dynamics of new accounts' behavior as well as Twitter's suspension policy prior to the recent change in ownership. This could stimulate further research to evaluate the impact of the ongoing changes introduced by the new administration.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.