Characterizing and automatically detecting crowdturfing in Fiverr and Twitter

As human computation on crowdsourcing systems has become popular and powerful for performing tasks, malicious users have started misusing these systems by posting malicious tasks, propagating manipulated contents, and targeting popular web services such as online social networks and search engines. Recently, these malicious users moved to Fiverr, a fast growing micro-task marketplace, where workers can post crowdturfing tasks (i.e., astroturfing campaigns run by crowd workers) and malicious customers can purchase those tasks for only $5. In this manuscript, we present a comprehensive analysis of crowdturfing in Fiverr and Twitter and develop predictive models to detect and prevent crowdturfing tasks in Fiverr and malicious crowd workers in Twitter. First, we identify the most popular types of crowdturfing tasks found in Fiverr and conduct case studies for these crowdturfing tasks. Second, we build crowdturfing task detection classifiers to filter these tasks and prevent them from becoming active in the marketplace. Our experimental results show that the proposed classification approach effectively detects crowdturfing tasks, achieving 97.35 % accuracy. Third, we analyze the real-world impact of crowdturfing tasks by purchasing active Fiverr tasks and quantifying their impact on a target site (Twitter). As part of this analysis, we show that current security systems inadequately detect crowd-sourced manipulation, which confirms the necessity of our proposed crowdturfing task detection approach. Finally, we analyze the characteristics of paid Twitter workers, find distinguishing features between these workers and legitimate Twitter accounts, and use these features to build classifiers that detect Twitter workers. Our experimental results show that our classifiers are able to detect Twitter workers effectively, achieving 99.29 % accuracy.