Fame for sale: Efficient detection of fake Twitter followers

Abstract

Fake followers are those Twitter accounts specifically created to inflate the number of followers of a target account. Fake followers are dangerous for the social platform and beyond, since they may alter concepts like popularity and influence in the Twittersphere-hence impacting on economy, politics, and society. In this paper, we contribute along different dimensions. First, we review some of the most relevant existing features and rules (proposed by Academia and Media) for anomalous Twitter accounts detection. Second, we create a baseline dataset of verified human and fake follower accounts. Such baseline dataset is publicly available to the scientific community. Then, we exploit the baseline dataset to train a set of machine-learning classifiers built over the reviewed rules and features. Our results show that most of the rules proposed by Media provide unsatisfactory performance in revealing fake followers, while features proposed in the past by Academia for spam detection provide good results. Building on the most promising features, we revise the classifiers both in terms of reduction of overfitting and cost for gathering the data needed to compute the features. The final result is a novel Class A classifier, general enough to thwart overfitting, lightweight thanks to the usage of the less costly features, and still able to correctly classify more than 95% of the accounts of the original training set. We ultimately perform an information fusion-based sensitivity analysis, to assess the global sensitivity of each of the features employed by the classifier.The findings reported in this paper, other than being supported by a thorough experimental methodology and interesting on their own, also pave the way for further investigation on the novel issue of fake Twitter followers.Keywords: Twitter, fake followers, anomalous account detection, baseline dataset, machine learning ✩ An extended version of this preprint has been accepted for publication on Elsevier's journal Decision Support Systems http: // dx.doi.org/ 10.1016/ j.dss.2015.09.003 . Please, cite this work as: "Cresci, S., Di Pietro, R., Petrocchi, M., Spognardi, A., & Tesconi, M. (2015

Arming the public with artificial intelligence to counter social bots

“…Social bot developers can populate bot profiles by searching and collecting material from other platforms. A more extreme example of these kinds of bots are identity thieves: they copy usernames, profile information, and pictures of other accounts and use them as their own, making only small changes (Cresci, Di Pietro, Petrocchi, Spognardi, & Tesconi, 2015). State-of-the-art machine learning technologies can be employed as part of the algorithms that automatically generate bot content (Freitas, Benevenuto, Ghosh, & Veloso, 2015).…”

Marketing through Instagram influencers: the impact of number of followers and product divergence on brand attitude

“…Thus, highly positive ratios could be an indication of true opinion leadership. On the other hand, it is said that users with a lot of followers in combination with only a few followees, are no "true" Twitter users (Siegler, 2009) or it indicate that the followers are artificially collected or "fake" (Cresci, Di Pietro, Petrocchi, Spognardi and Tesconi, 2015). In contrast, a user with many followees has more opportunities to learn about different topics and opinions, and thus more ability to look beyond their own social environment, which might be beneficial in terms of opinion leadership (Williams 2006).…”

The Paradigm-Shift of Social Spambots

“…Fake followers are another kind of malicious accounts that recently gained interest both from platform administrators and from the scientific world [12]. Given that fake followers are rather simplistic in their design and functioning, they can serve as a weak baseline against which to compare social spambots.…”