Identifying the Community Roles of Social Capitalists in the Twitter Network

TL;DR

This paper analyzes the distribution and roles of social capitalists on Twitter, revealing their high visibility and strategic positions within community structures, which may facilitate malicious influence.

Contribution

The study extends existing community role measures with a new unsupervised approach, providing deeper insights into social capitalists' positions in Twitter networks.

Findings

Social capitalists occupy highly visible community roles

They are strategically positioned across multiple communities

Their roles facilitate increased influence and visibility

Abstract

In the context of Twitter, social capitalists are specific users trying to increase their number of followers and interactions by any means. These users are not healthy for the Twitter network since they flaw notions of influence and visibility. Indeed, it has recently been observed that they are real and active users that can help malicious users such as spammers gaining influence. Studying their behavior and understanding their position in Twitter is thus of important interest. A recent work provided an efficient way to detect social capitalists using two simple topological measures. Based on this detection method, we study how social capitalists are distributed over Twitter's friend-to-follower network. We are especially interested in analyzing how they are organized, and how their links spread across the network. Answering these questions allows to know whether the social capitalism methods increase the actual visibility on the service. To that aim, we study the position of social capitalists on Twitter w.r.t. the community structure of the network. We base our work on the concept of community role of a node, which describes its position in a network depending on its connectivity at the community level. The topological measures originally defined to characterize these roles consider only some aspects of community-related connectivity and rely on a set of empirically fixed thresholds. We first show the limitations of such measures and then extend and generalize them by considering new aspects of the community-related connectivity. Moreover, we use an unsupervised approach to distinguish the roles, in order to provide more flexibility relatively to the studied system. We then apply our method to the case of social capitalists and show that they are highly visible on Twitter, due to the specific roles they occupy.