Analyzing and Detecting Collusive Users Involved in Blackmarket Retweeting Activities

Abstract

With the rise in popularity of social media platforms like Twitter, having higher influence on these platforms has a greater value attached to it, since it has the power to influence many decisions in the form of brand promotions and shaping opinions. However, blackmarket services that allow users to inorganically gain influence are a threat to the credibility of these social networking platforms. Twitter users can gain inorganic appraisals in the form of likes, retweets, and follows through these blackmarket services either by paying for them or by joining syndicates wherein they gain such appraisals by providing similar appraisals to other users. These customers tend to exhibit a mix of organic and inorganic retweeting behavior, making it tougher to detect them. In this article, we investigate these blackmarket customers engaged in collusive retweeting activities. We collect and annotate a novel dataset containing various types of information about blackmarket customers and use these sources of information to construct multiple user representations. We adopt Weighted Generalized Canonical Correlation Analysis (WGCCA) to combine these individual representations to derive user embeddings that allow us to effectively classify users as: genuine users, bots, promotional customers, and normal customers. Our method significantly outperforms state-of-the-art approaches (32.95% better macro F1-score than the best baseline).

Who pushes the discussion on wind energy? An analysis of self-reposting behaviour on Twitter

“…However, some bad practices have emerged, representing a threat to the credibility of these social networking platforms. For example, Arora et al (2020) has investigated collusive retweeting activities, e.g., those services provided by paying for them to gain influence inorganically. The impact of spammers on Twitter networks has instead been addressed by Fronzetti Colladon and Gloor (2019), while Liu et al (2019) has analysed the crowd-retweeting spam.…”

Weakening the Inner Strength: Spotting Core Collusive Users in YouTube Blackmarket Network

“…further studied the working of premium and freemium blackmarket services in providing collusive appraisals on Twitter. Arora et al (2020) further investigated the blackmarket customers engaged in collusive retweeting activities using a multiview learning based approach. Chetan et al (2019) proposed CoReRank, an unsupervised method to detect collusive retweeters and suspicious tweets on Twitter.…”

ABOME: A Multi-platform Data Repository of Artificially Boosted Online Media Entities

“…(Arora, Paka, and Chakraborty 2019) detected tweets submitted to blackmarket services using a multitask learning approach. (Arora et al 2020) proposed a multi-view learning-based approach to detect collusive retweeters by utilizing various attribute and network views based on the user's posts and interactions on the social graph. We encourage the reader to go through for a comprehensive survey on analyzing and detecting collusive activities in online media platforms.…”

“…However, there are relatively fewer studies on the detection and analysis of collusive activities that result in an artificial boosting of social growth. Our recent investigations (Chetan et al 2019;Dutta et al 2018;Dutta and Chakraborty 2019;Arora, Paka, and Chakraborty 2019;Dhawan et al 2019;Arora et al 2020;Sankar Dutta et al 2020) revealed that existing fraud detection strategies are not suitable for blackmarket-driven collusive entity detection. These studies reported that collusive users are not bots or fake users; rather, they are normal users showing a mix of organic and inorganic activities with no synchronicity across their behaviors.…”