Twitter trolls: a linguistic profile of anti-democratic discourse

Abstract

This article focuses on anti-democratic discourse and investigates the linguistic profile of Twitter trolls. The troll data consist of some 3.5 million messages in English obtained through Twitter in late 2018. These data originate from potentially state-backed information operations aimed at sowing discord in Western societies. The baseline data, against which the troll data are compared, contain circa 4.4 million messages in English drawn from the Nordic Tweet Stream corpus. A machine learning application that enables us to select genuine personal messages in this corpus is used to prune the data. The empirical part investigates frequencybased characteristics of the two datasets. We utilize a set of automatically-extracted word-list information and the observed frequencies of personal pronouns. Our empirical findings show considerable quantitative differences so that the troll data are shorter, make use of a smaller number of lexical types and tokens, and resemble more formal registers, while the personal messages are more spoken-like. The results could be used to improve automated detection systems whose purpose is to identify troll accounts.

Early detection of internet trolls: Introducing an algorithm based on word pairs / single words multiple repetition ratio

“…More importantly, it is clear that as long as we choose different baseline datasets, we will most likely get different results. Thus, in stark contrast to what Zannettou et al discovered, Lundberg and Laitinen in [16] argued that the troll messages are shorter than their baseline material and resemble more formal registers in that they contain a higher proportion of nouns. As we will see below, the latter observation is more compatible with our own results than the former.…”

The Impact of Culture on Online Toxic Disinhibition: Trolling in India and the USA

“…Because political trolling is widespread on the platform, recent scholarly attention was given to the automatic identification of online trolling on Twitter through linguistic analysis of trolling datasets [e.g., 5,20,21,22,23]. Efforts to identify trolling on Twitter are important, as these trolls negatively affect the political process, cause distrust in the political systems, and increase political polarization and conflict [24].…”

“…The "Kremlin trolls" simply aim to undermine American political institutions, spread discord within American society, and undermine the United States' global influence [25]. Scholars [23] focused on anti-democratic discourse and investigated the linguistic profile of Twitter trolls and found that trolls tend to write shorter posts and used a smaller number of lexical types and tokens. Others [22] developed machine learning models that predict whether a Twitter account is a Russian troll, using both behavioral and linguistic features.…”

Do I sound American? How message attributes of IRA disinformation relate to Twitter engagement

“…Troll messages are "powerful weapon in modern hybrid warfare" (Monakhov, 2020, p. 2). For state-backed anti-democratic agents seeking to shift power dynamics, language can be a crucial resource for the strategic amplification of messages (Lundberg & Laitinen, 2020). Topics and sentiments that accompany words operate simultaneously in a message.…”