Detecting Marionette Microblog Users for Improved Information Credibility

Abstract

In this paper, we propose to detect a special group of microblog users: the "marionette" users, who are created or employed by backstage "puppeteers", either through programs or manually. Unlike normal users that access microblog for information sharing or social communication, the marionette users perform specific tasks to earn financial profits. For example, they follow certain users to increase their "statistical popularity", or retweet some tweets to amplify their "statistical impact". The fabricated follower or retweet counts not only mislead normal users to wrong information, but also seriously impair microblog-based applications, such as hot tweets selection and expert finding. In this paper, we study the important problem of detecting marionette users on microblog platforms. This problem is challenging because puppeteers are employing complicated strategies to generate marionette users that present similar behaviors as normal users. To tackle this challenge, we propose to take into account two types of discriminative information: 1) individual user tweeting behavior and 2) the social interactions among users. By integrating both information into a semi-supervised probabilistic model, we can effectively distinguish marionette users from normal ones. By applying the proposed model to one of the most popular microblog platforms (Sina Weibo) in China, we find that the model can detect marionette users with F -measure close to 0.9. In addition, we apply the proposed model to calculate the marionette ratio of the top 200 most followed microbloggers and the top 50 most retweeted posts in Sina Weibo. To accelerate the detecting speed and reduce feature generation cost, we further propose a light-weight model which utilizes fewer features to identify marionettes from retweeters.

Credibility in Online Social Networks: A Survey

“…Consequently, misinformation may spread widely and can cause a lot of damage before human operators detect it. However, others [84], [109]- [118] have stressed that manual methods of credibility assessment are more accurate than automation-based methods. They used statistical analyses to interpret results and measure user behavior regarding particular news.…”

“…Cognitive ability can be defined ''psychological function of knowledge that consists of elements like vigilance, sensory system, learning, and thought processes'' [109], correlating exactly which concerns are being regarded, including beliefs, reasons, intuitions, and expertise [110]. Beyond social network analysis, cognitive psychology helps to identify the mental ability tasked with data dissemination [111].…”

“…This method addresses four main questions: message consistency, message coherence, source credibility, and general message acceptability. Many researchers [84], [109]- [118] using this approach employ human perceptions and knowledge to evaluate fake content and misinformation in tweets.…”

A Spammer Identification Method for Class Imbalanced Weibo Datasets

“…They considered that there may be only a small amount of labeled data and a large amount of unlabeled data in many scenarios. At present, the features used to distinguish between spammers and non-spammers mainly include account-based features [3], content-based features [24], relationship(graph)based features [5], [28], behavior-based features [25], [27], time-based features [29], and hybrid features [1], [2].…”

Knowledge discovery from evolving social media data

“…In addition, most existing methods (e.g. Wu et al [2013]) consider the typical and out-dated model of a fraudster who has uniform posting frequency and a followers-to-followees ratio close to 1 -nowadays, many fraudsters are more sophisticated. Ghosh et al [2011] address a problem similar to ours, but use the URLs found in tweets instead of retweet threads in conjunction with a time and user-based entropy to classify posting activity and content.…”